Cartilage and Osteoarthritis

M E T H O D S I N M O L E C U L A R M E D I C I N E TM
Cartilage and
Osteoarthritis
Volume 1
Cellular and Molecular Tools
Edited by
Massimo Sabatini
Philippe Pastoureau
Frédéric De Ceuninck
Cartilage and Osteoarthritis
M E T H O D S I N M O L E C U L A R M E D I C I N E™
John M. Walker, SERIES EDITOR

111.
111 Chemosensitivity: Volume 2, In Vivo Mod- 95.
95 Hepatitis B and D Protocols: Volume 1,
els, Imaging, and Molecular Regulators, ed- Detection, Genotypes, and Characterization,
ited by Rosalyn D. Blumethal, 2005 edited by Robert K. Hamatake and Johnson Y.
110.
110 Chemosensitivity: Volume 1, In Vitro As- N. Lau, 2004
says, edited by Rosalyn D. Blumethal, 2005 94.
94 Molecular Diagnosis of Infectious Diseases,
109.
109 Adoptive Immunotherapy, Methods and Second Edition, edited by Jochen Decker and
Protocols, edited by Burkhard Ludewig and Udo Reischl, 2004
Matthias W. Hoffman, 2005 93.
93 Anticoagulants, Antiplatelets, and
108. Hypertension, Methods and Protocols, Thrombolytics, edited by Shaker A. Mousa,
edited by Jérôme P. Fennell and Andrew 2004
H. Baker, 2005 92.
92 Molecular Diagnosis of Genetic Diseases,
107.
107 Human Cell Culture Protocols, Second Second Edition, edited by Rob Elles and
Edition, edited by Joanna Picot, 2005 Roger Mountford, 2004
106.
106 Antisense Therapeutics, Second Edition, 91.
91 Pediatric Hematology: Methods and
edited by M. Ian Phillips, 2005 Protocols, edited by Nicholas J. Goulden and
105.
105 Developmental Hematopoiesis: Methods Colin G. Steward, 2003
and Protocols, edited by Margaret H. 90.
90 Suicide Gene Therapy: Methods and Reviews,
Baron, 2005 edited by Caroline J. Springer, 2004
104.
104 Stroke Genomics: Methods and Reviews, 89.
89 The Blood–Brain Barrier: Biology and
edited by Simon J. Read and David Virley, Research Protocols, edited by Sukriti Nag,
2004 2003
103.
103 Pancreatic Cancer: Methods and Protocols, 88.
88 Cancer Cell Culture: Methods and
edited by Gloria H. Su, 2004 Protocols, edited by Simon P. Langdon,
102.
102 Autoimmunity: Methods and Protocols, 2003
edited by Andras Perl, 2004 87.
87 Vaccine Protocols, Second Edition, edited by
101.
101 Cartilage and Osteoarthritis: Volume 2, Andrew Robinson, Michael J. Hudson, and
Structure and In Vivo Analysis, edited by Martin P. Cranage, 2003
Frédéric De Ceuninck, Massimo Sabatini, 86.
86 Renal Disease: Techniques and Protocols,
and Philippe Pastoureau, 2004 edited by Michael S. Goligorsky, 2003
100.
100 Cartilage and Osteoarthritis: Volume 1, 85.
85 Novel Anticancer Drug Protocols, edited by
Cellular and Molecular Tools, edited by John K. Buolamwini and Alex A. Adjei, 2003
Massimo Sabatini, Philippe Pastoureau, and
84.
84 Opioid Research: Methods and Protocols,
Frédéric De Ceuninck, 2004
edited by Zhizhong Z. Pan, 2003
99.
99 Pain Research: Methods and Protocols,
edited by David Z. Luo, 2004 83.
83 Diabetes Mellitus: Methods and Protocols,
edited by Sabire Özcan, 2003
98.
98 Tumor Necrosis Factor: Methods and
Protocols, edited by Angelo Corti and Pietro 82.
82 Hemoglobin Disorders: Molecular Methods
Ghezzi, 2004 and Protocols, edited by Ronald L. Nagel,
2003
97.
97 Molecular Diagnosis of Cancer: Methods
and Protocols, Second Edition, edited by 81.
81 Prostate Cancer Methods and Protocols,
Joseph E. Roulston and John M. S. Bartlett, edited by Pamela J. Russell, Paul Jackson,
2004 and Elizabeth A. Kingsley, 2003
96.
96 Hepatitis B and D Protocols: Volume 2, 80.
80 Bone Research Protocols, edited by Miep H.
Immunology, Model Systems, and Clinical Helfrich and Stuart H. Ralston, 2003
Studies, edited by Robert K. Hamatake and 79.
79 Drugs of Abuse: Neurological Reviews and
Johnson Y. N. Lau, 2004 Protocols, edited by John Q. Wang, 2003
M E T H O D S I N M O L E C U L A R M E D I C I N E™
Cartilage
and
Osteoarthritis
VOLUME 1
Edited by
Massimo Sabatini
Philippe Pastoureau
Division de Rhumatologie
Institut de Recherches Servier
Suresnes, France
© 2004 Humana Press Inc.
999 Riverview Drive, Suite 208
Totowa, New Jersey 07512
www.humanapress.com
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in
any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise
without written permission from the Publisher. Methods in Molecular MedicineTM is a trademark of The
Humana Press Inc.
All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not
necessarily reflect the views of the publisher.
This publication is printed on acid-free paper. ∞

ANSI Z39.48-1984 (American Standards Institute)
Permanence of Paper for Printed Library Materials.
Cover design by Patricia F. Cleary.

Cover illustrations: Stained sections of normal (background) and osteoarthritic cartilage (foreground).
Artwork courtesy of Massimo Sabatini, Philippe Pastoureau, and Frédéric De Ceuninck.
For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact
Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341;
E-mail: humana@humanapr.com; or visit our Website: www.humanapress.com
Photocopy Authorization Policy:
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific
clients, is granted by Humana Press Inc., provided that the base fee of US $25.00 per copy is paid directly
to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations
that have been granted a photocopy license from the CCC, a separate system of payment has been arranged
and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is:
[1-58829-247-9/04 $25.00].
Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
eISBN 1-59259-810-2
ISSN 1543-1894
Library of Congress Cataloging-in-Publication Data

Cartilage and osteoarthritis.
p. ; cm. — (Methods in molecular medicine ; 100, etc.)
Contents: v. 1. Cellular and molecular tools / edited by Massimo Sabatini, Philippe Pastoureau,
Frédéric De Ceuninck.
Includes bibliographical references and index.
ISBN 1-58829-247-9 (v. 1 : alk. paper)
1. Cartilage cells—Laboratory manuals. 2. Osteoarthritis—Laboratory manuals. I. Sabatini,
Massimo. II. Pastoureau, Philippe. III. De Ceuninck, Frédéric. IV. Series.
[DNLM: 1. Chondrocytes—pathology. 2. Osteoarthritis. 3. Chondrocytes—metabolism. WE 348
C3268 2004]
QP88.23.C36 2004
616.7'223—dc22
200302809
Preface
Osteoarthritis (OA), the most common form of arthritis, is generally
characterized by a slowly progressive degeneration of articular cartilage,
particularly in the weight-bearing joints. It has a stronger prevalence in women,
and its incidence increases with age. OA is a major and growing health concern in
developed countries, owing to steadily increasing life expectancy and the demand
for better quality of life. Because of its chronic nature and nonfatal outcome, OA
affects the growing population of the elderly over an increasing time span.
Moreover, despite its relatively benign character, OA is one of the most disabling
diseases; it is responsible for increasing financial and social burdens in terms of
medical treatments, forced inactivity, loss of mobility, and dependence.
Despite a growing awareness of OA as a medical problem that has yet to reach
its maximum impact on society, there is a surprising absence of effective medical
treatments beyond pain control and surgery. So far, only symptom-modifying drugs
are available, while there remains a major demand for disease-modifying treatments
of proven clinical efficacy. This demand will hopefully be met in the future by
some of the drugs that have been pressed into development and are now at different
stages of clinical investigation. Nevertheless, the current lack of effective treatments
reflects a still insufficient knowledge of cartilage with respect to its metabolism,
interactions with other joint tissues, and causes and mechanisms (possibly of very
different nature) leading to failure of its turnover. As is seen in other therapeutic
fields, the future availability of better drugs will depend on a deeper knowledge of
OA physiopathology, allowing rational definition of new molecular targets for
pharmacological intervention. This new interest in OA is fostering an intense
research effort both in academic institutions and in the pharmaceutical industry.
In this context, two volumes of the Methods in Molecular Medicine™ series
are dedicated to research protocols on cartilage and osteoarthritis. Cartilage
and Osteoarthritis, Volume 1, Cellular and Molecular Tools combines
classical but still evolving techniques with emerging methods that promise to
add critical knowledge to our understanding of cartilage metabolism in health
and disease. Authors with hands-on expertise have described protocols for the
in vitro study of normal and osteoarthritic cartilage through biochemical,
biomolecular, immunological, and physical approaches. Volume 2: Structure
and In Vivo Analysis is dedicated to procedures for study at the tissue level of
turnover, structure, and functioning of normal and diseased articular cartilage,
through invasive and noninvasive means.
v
vi Preface
We hope and expect that the two volumes of Cartilage and Osteoarthritis
will constitute a welcome addition to the literature of research protocols, as
well as a helpful and trusted laboratory companion.
Massimo Sabatini
Philippe Pastoureau
Acknowledgments
We would like to thank Corinne Morlat, Research Assistant, Institut de
Recherches Servier, for her invaluable secretarial help in editing this book.
Contents of Volume 1
Preface .............................................................................................................. v
Contents of Volume 2 ...................................................................................... ix
Contributors ..................................................................................................... xi
1 Culture and Phenotyping of Chondrocytes in Primary Culture
Sylvie Thirion and Francis Berenbaum ................................................. 1
2 Culture of Chondrocytes in Alginate Beads
Frédéric De Ceuninck, Christophe Lesur, Philippe Pastoureau,
Audrey Caliez, and Massimo Sabatini ............................................ 15
3 Immortalization of Human Articular Chondrocytes
for Generation of Stable, Differentiated Cell Lines
Mary B. Goldring ................................................................................ 23
4 Culture of Immortalized Chondrocytes and Their Use As Models
of Chondrocyte Function
Mary B. Goldring ................................................................................ 37
5 Generation of Pluripotent Stem Cells and Their Differentiation
to the Chondrocytic Phenotype
Luis A. Solchaga, Jean F. Welter, Donald P. Lennon,
and Arnold I. Caplan ...................................................................... 53
6 Semiquantitative Analysis of Gene Expression in Cultured
Chondrocytes by RT-PCR
Gaëlle Rolland-Valognes ..................................................................... 69
7 Quantification of mRNA Expression Levels
in Articular Chondrocytes With PCR Technologies
Audrey McAlinden, Jochen Haag, Brigitte Bau,
Pia M. Gebhard, and Thomas Aigner ............................................. 79
8 RNA Extraction From Cartilage
Frédéric Mallein-Gerin and Jérôme Gouttenoire ............................. 101
9 Gene Expression Analysis in Cartilage by In Situ Hybridization
Frédéric Mallein-Gerin and Jérôme Gouttenoire ............................. 105
10 Analysis of Differential Gene Expression in Healthy
and Osteoarthritic Cartilage and Isolated Chondrocytes
by Microarray Analysis
Thomas Aigner, Joachim Saas, Alexander Zien, Ralf Zimmer,
Pia M. Gebhard, and Thomas Knorr ............................................. 109
11 High-Efficiency Nonviral Transfection of Primary Chondrocytes
Jean F. Welter, Luis A. Solchaga, and Matthew C. Stewart ............. 129
vii
viii Contents of Volume 1
12 In Vitro Gene Transfer to Chondrocytes and Synovial Fibroblasts

by Adenoviral Vectors
Jean-Noel Gouze, Martin J. Stoddart, Elvire Gouze,
Glyn D. Palmer, Steven C. Ghivizzani, Alan J. Grodzinsky,
and Christopher H. Evans ............................................................. 147
13 Changes of Chondrocyte Metabolism In Vitro:
An Approach by Proteomic Analysis
Anne-Marie Freyria and Michel Becchi ............................................ 165
14 Analysis of Chondrocyte Functional Markers
and Pericellular Matrix Components by Flow Cytometry
Gust Verbruggen, Jun Wang, Lai Wang, Dirk Elewaut,
and Eric M. Veys ........................................................................... 183
15 A Simple and Reliable Assay of Proteoglycan Synthesis
by Cultured Chondrocytes
Frédéric De Ceuninck and Audrey Caliez ........................................ 209
16 Assays of Proteoglycan and Collagen Degradation
in Cultures of Rabbit Cartilage Explants
Christophe Lesur and Massimo Sabatini ........................................... 219
17 Production of Antibodies Against Degradative Neoepitopes
in Aggrecan
John S. Mort and Peter J. Roughley .................................................. 237
18 Immunoassays for Collagens in Chondrocyte
and Cartilage Explant Cultures
R. Clark Billinghurst, Fackson Mwale, Anthony Hollander,
Mirela Ionescu, and A. Robin Poole ............................................. 251
19 Detection of Apoptosis in Cartilage In Situ and in Isolated
Chondrocytes
Darryl D. D’Lima, Klaus Kuhn, and Martin K. Lotz .......................... 275
20 Expression, Activity, and Regulation of MAP Kinases in Cultured
Chondrocytes
Jang-Soo Chun ................................................................................... 291
21 Mechanical Loading of Chondrocytes Embedded in 3D Constructs:
In Vitro Methods for Assessment of Morphological and Metabolic
Response to Compressive Strain
David A. Lee and Martin M. Knight .................................................. 307
22 In Vitro Physical Stimulation of Tissue-Engineered
and Native Cartilage
Kelvin W. Li, Travis J. Klein, Kanika Chawla, Gayle E. Nugent,
Won C. Bae, and Robert L. Sah .................................................... 325
Index ............................................................................................................ 353
CONTENTS OF COMPANION VOLUME
Volume 2: Structure and In Vivo Analysis
1 Generation and Use of Transgenic Mice As Models of Osteoarthritis

Anna-Marja Säämänen and Eero Vuorio
2 Development and Clinical Application in Arthritis of a New
Immunoassay for Serum Type IIA Procollagen NH2 Propeptide
Jean-Charles Rousseau, Linda J. Sandell, Pierre D. Delmas,
and Patrick Garnero
3 Histological and Immunohistological Studies on Cartilage
James Melrose, Susan Smith, and Peter Ghosh
4 Histochemical Visualization of the Cartilage Hyaladherins Using
a Biotinylated Hyaluronan Oligosaccharide Bioaffinity Probe
James Melrose and Susan Smith
5 Methods for Cartilage and Subchondral Bone Histomorphometry
Philippe Pastoureau and Agnès Chomel
6 Laser-Mediated Microdissection As a Tool for Molecular Analysis
in Arthritis
Martin Judex, Elena Neumann, Steffen Gay, and Ulf Müller-Ladner
7 Analysis of Protein Distribution in Cartilage Using
Immunofluorescence and Laser Confocal Scanning Microscopy
Stephan Soeder, Alexander Kuhlmann, and Thomas Aigner
8 Molecular and Biochemical Assays of Cartilage Components
Caroline D. Hoemann
9 Mechanical Characterization of Native
and Tissue-Engineered Cartilage
Albert C. Chen, Stephen M. Klisch, Won C. Bae,
Michele M. Temple, Kevin B. McGowan, Kenneth R. Gratz,
Barbara L. Schumacher, and Robert L. Sah
10 Noninvasive Study of Human Cartilage Structure by MRI
Felix Eckstein
11 High-Resolution MRI Techniques for Studies in Small-Animal Models
Olivier Beuf
12 High-Resolution Imaging of Osteoarthritis Using Microcomputed
Tomography
Lydia Wachsmuth and Klaus Engelke
ix
x Contents of Volume 2
13 High-Resolution Ultrasonography for Analysis

of Age- and Disease-Related Cartilage Changes
Amena Saïed and Pascal Laugier
14 Evaluation of Cartilage Composition and Degradation
by High-Resolution Magic-Angle Spinning Nuclear
Magnetic Resonance
Jürgen Schiller, Daniel Huster, Beate Fuchs, Lama Naji,
Jörn Kaufmann, and Klaus Arnold
15 Pulsed-Field Gradient–Nuclear Magnetic Resonance
(PFG NMR) to Measure the Diffusion of Ions and Polymers
in Cartilage: Applications in Joint Diseases
Jürgen Schiller, Lama Naji, Robert Trampel, Wilfred Ngwa,
Robert Knauss, and Klaus Arnold
16 Dynamics of Collagen in Articular Cartilage Studied
by Solid-State NMR Methods
Daniel Huster, Jürgen Schiller, and Klaus Arnold
17 Computerized Protocol for Anatomical and Functional Studies
of Joints
Sandra Martelli and Stefano Zaffagnini
Contributors
THOMAS AIGNER • Cartilage Research, Department of Pathology, University
of Erlangen-Nürnberg, Erlangen, Germany
WON C. BAE • Department of Bioengineering and Whitaker Institute
of Biomedical Engineering, University of California, San Diego, San
Diego, CA
BRIGITTE BAU • Cartilage Research, Department of Pathology, University
MICHEL BECCHI • UMR 5086 CNRS, Institut de Biologie et Chimie des
Protéines, Université Lyon I et IFR 128-Biosciences Lyon-Gerland, Lyon,
France
FRANCIS BERENBAUM • UPRES-A CNRS 7079, Université Pierre
et Marie Curie, Paris, France
R. CLARK BILLINGHURST • Department of Health Sciences, St. Lawrence
College, Kingston, Ontario, Canada
AUDREY CALIEZ • Division de Rhumatologie, Institut de Recherches Servier,
Suresnes, France
ARNOLD I. CAPLAN • Skeletal Research Center, Department of Biology, Case
Western Reserve University, Cleveland, OH
KANIKA CHAWLA • Department of Bioengineering and Whitaker Institute
Diego, CA
JANG-SOO CHUN • Department of Life Science, Kwangju Institute
of Science and Technology, Gwangju, Korea
FRÉDÉRIC DE CEUNINCK • Division de Rhumatologie, Institut de Recherches
Servier, Suresnes, France
DARRYL D. D’LIMA • Director, Orthopaedic Research Laboratories, Scripps
Clinic Center for Orthopaedic Research and Education, La Jolla, CA
DIRK ELEWAUT • Department of Rhumatology, Ghent University Hospital,
Ghent University, Ghent, Belgium
CHRISTOPHER H. EVANS • Center for Molecular Orthopaedics, Harvard
Medical School, Boston, MA
ANNE-MARIE FREYRIA • UMR 5086 CNRS, Institut de Biologie et Chimie
des Protéines, Université Lyon I et IFR 128-Biosciences Lyon-Gerland,
Lyon, France
PIA M. GEBHARD • Cartilage Research, Department of Pathology, University
xi
xii Contributors
STEVEN C. GHIVIZZANI • Center for Molecular Orthopedics, Harvard Medical

School, Boston, MA
MARY B. GOLDRING • Beth Israel Deaconess Medical Center, Division
of Rheumatology, New England Baptist Bone and Joint Institute,
Harvard Institutes of Medicine, Boston, MA
JÉRÔME GOUTTENOIRE • Laboratory of Biology and Engineering of Cartilage,
CNRS UMR 5086, Institut de Biologie et Chimie des Protéines, Lyon, France
ELVIRE GOUZE • Center for Molecular Orthopaedics, Harvard Medical
School, Boston, MA
JEAN-NOEL GOUZE • Center for Molecular Orthopaedics, Harvard Medical
School, Boston, and Center for Biomedical Engineering, Massachusetts
Institute of Technology, Cambridge, MA
ALAN J. GRODZINSKY • Center for Biomedical Engineering, Massachusetts
Institute of Technology, Cambridge, MA
JOCHEN HAAG • Cartilage Research, Department of Pathology, University
ANTHONY HOLLANDER • University of Bristol Academic Rheumatology,
AMBI, Avon Orthopaedic Centre, Southmead Hospital, Bristol, United
Kingdom
MIRELA IONESCU • Joint Diseases Laboratory, Shriners Hospital
for Children, Montreal, Quebec, Canada
TRAVIS J. KLEIN • Department of Bioengineering and Whitaker Institute
Diego, CA
MARTIN M. KNIGHT • Department of Engineering, Queen Mary College,
University of London, London, United Kingdom
THOMAS KNORR • Cartilage Research, Department of Pathology, University
of Erlangen–Nürnberg, Erlangen, Germany
KLAUS KUHN • Division of Arthritis Research, The Scripps Research
Institute, La Jolla, CA
DAVID A. LEE • IRC in Biomedical Materials and Medical Engineering
Division, Department of Engineering, Queen Mary College, University
of London, London, United Kingdom
DONALD P. LENNON • Skeletal Research Center, Department of Biology, Case
Western Reserve University, Cleveland, OH
CHRISTOPHE LESUR • Division de Rhumatologie, Institut de Recherches
KELVIN W. LI • Department of Bioengineering and Whitaker Institute
Diego, CA
Contributors xiii
MARTIN K. LOTZ • Division of Arthritis Research, The Scripps Research

Institute, La Jolla, CA
FRÉDÉRIC MALLEIN-GERIN • Laboratory of Biology and Engineering of Cartilage,
CNRS UMR 5086, Institut de Biologie et Chimie des Protéines, Lyon, France
AUDREY MCALINDEN • Department of Orthopaedic Surgery, School
of Medicine, Washington University in St. Louis, St. Louis, MO
J OHN S. M ORT • Joint Diseases Laboratory, Shriners Hospital
for Children, and Departments of Surgery and Medicine, McGill
University, Montreal, Canada
FACKSON MWALE • The Sir Mortimer B. Davis-Jewish General Hospital,
Montreal, Quebec, Canada
GAYLE E. NUGENT • Department of Bioengineering and Whitaker Institute
of Biomedical Engineering, University of California, San Diego,
San Diego, CA
GLYN D. PALMER • Center for Molecular Orthopaedics, Brigham
and Women's Hospital, Harvard Medical School, Boston, MA
PHILIPPE PASTOUREAU • Division de Rhumatologie, Institut
de Recherches Servier, Suresnes, France
A. ROBIN POOLE • Joint Diseases Laboratory, Shriners Hospital
for Children; McGill University, Montreal, Quebec, Canada
GAËLLE ROLLAND-VALOGNES • Division de Rhumatologie, Institut
de Recherches Servier, Suresnes, France
PETER J. ROUGHLEY • Genetics Unit, Shriners Hospital for Children,
and Departments of Surgery and Human Genetics, McGill University,
Montreal, Canada
JOACHIM SAAS • Disease Group Osteoarthritis, Aventis Pharma GmbH,
Frankfurt, Germany
MASSIMO SABATINI • Division de Rhumatologie, Institut de Recherches
ROBERT L. SAH • Department of Bioengineering and Whitaker Institute
of Biomedical Engineering, University of California, San Diego,
San Diego, CA
LUIS A. SOLCHAGA • Department of Orthopaedics, Case Western Reserve
University and University Hospitals of Cleveland, Cleveland, OH
MATTHEW C. STEWART • Department of Veterinary Clinical
Medicine, University of Illinois at Urbana–Champaign, Urbana, IL
MARTIN J. STODDART • Center for Molecular Orthopaedics, Harvard Medical
School, Boston, MA, and Laboratory for Experimental Cartilage
Research, Center for Rheumatology and Bone Disease, Zürich,
Switzerland
xiv Contributors
SYLVIE THIRION • UMR CNRS 6544, Laboratoire Interactions Cellulaires
Neuroendocriennes, Institut Fédératif de Recherche Jean Roche, Faculté
de Médecine Nord, Université Aix–Marseille II, Marseille, France
GUST VERBRUGGEN • Department of Rhumatology, Ghent University
Hospital, Ghent University, Ghent, Belgium
ERIC M. VEYS • Department of Rhumatology, Ghent University Hospital,
Ghent University, Ghent, Belgium
JUN WANG • Department of Rhumatology, Ghent University
LAI WANG • Department of Rhumatology, Ghent University
JEAN F. WELTER • Department of Orthopaedics, Case Western Reserve
University and University Hospitals of Cleveland, Cleveland, OH
ALEXANDER ZIEN • Cartilage Research, Department of Pathology, University
RALF ZIMMER • Cartilage Research, Department of Pathology, University
Primary Culture of Chondrocytes 1
1
Culture and Phenotyping
of Chondrocytes in Primary Culture
Sylvie Thirion and Francis Berenbaum
Summary
The culture of chondrocytes is one of the most powerful tool for exploring the intracellular
and molecular features of chondrocyte differentiation and activation. However, chondrocytes
tend to dedifferentiate to fibroblasts when they are subcultured, which is a major problem. This
chapter describes several protocols for culturing chondrocytes of different anatomical origins
(articular and costal chondrocytes) from various species (humans, mice, rabbits, and cattle).
All these protocols involve primary cultures in order to limit dedifferentiation. This chapter
also describes a new protocol for culturing mouse articular chondrocytes.
Key Words: Primary cell culture; isolation; cartilage; articular chondrocytes; costal
chondrocytes; human; mice; rabbit; cattle.
1. Introduction
Normal cartilage has two main components: the collagen- and proteoglycan-
rich extracellular matrix and a population of isolated chondrocytes lying within
this matrix. From the outermost cartilage layer to the growth plate zone, these
chondrocytes show phenotypic variations that reflect their progress through a
cascade of differentiating events triggered by environmental signals that either
stimulate or suppress the conversion from articular chondrocytes to hyper-
trophic chondrocytes (also called epiphyseal chondrocytes) (1–3). Articular
chondrocytes produce a matrix that confers tensile strength and flexibility to
articular surfaces, whereas growth plate chondrocytes produce a matrix capa-
ble of undergoing mineralization (4). These functional differences explain the
specific phenotype of articular and hypertrophic chondrocytes (2). Articular
chondrocytes mainly express collagen types II, IX, and XI, as well as aggrecan
(1,5). Hypertrophic chondrocytes reach more advanced stages of differentia-
From: Methods in Molecular Medicine, Vol. 100: Cartilage and Osteoarthritis, Vol. 1: Cellular and Molecular Tools
Edited by: M. Sabatini, P. Pastoureau, and F. De Ceuninck © Humana Press Inc., Totowa, NJ
1
2 Thirion and Berenbaum
tion and express collagen type X and alkaline phosphatase, along with col-
lagen types II and IX and aggrecan (6–8).
Chondrocyte cultures remain one of the most powerful tools for investigat-
ing intracellular and molecular events associated with chondrocyte differentia-
tion and activation (9,10). However, culturing can produce artifacts that bias
results (1,2). The main problem is that cultured chondrocytes tend to dediffer-
entiate into fibroblasts. Factors associated with an increased tendency toward
dedifferentiation include the following:
1. Low-density plating (11).
2. Monolayer culturing (3,12).
3. Treatment with proinflammatory cytokines, such as interleukin-1 (IL-1) (13–15).
4. Extraction from human adult cartilage.
5. Extraction from chondrosarcoma or immortalization (16–18).
Conversion from the chondrocyte to the fibroblast phenotype can be detected
based on a switch from collagen type II (articular chondrocytes) or type X (epi-
physeal chondrocytes) to collagen types I and III, and from high-molecular-
weight proteoglycans (aggrecan) to low-molecular-weight proteoglycans
(biglycan and decorin) (13–15). To minimize conversion to fibroblasts, studies
of nonsubcultured chondrocytes in monolayers have made extensive use of cells
from young animals such as rabbits, cattle, or rats, which are more stable than
human adult chondrocytes.
Over the last 20 yr, efforts to obtain a better phenotype than that of 2D cul-
tured chondrocytes have involved embedding the cells in an artificial matrix
made of alginate (see Chapter 2), agarose, or collagens (3,12,16). However,
although this approach improves the chondrocyte phenotype, it slows cell
growth, so that less material is available for study. In addition, extracting the
cells from the matrix is technically challenging. Cell lines from chondrosar-
coma produce larger numbers of cells but have tumorigenic properties that may
fail to replicate physiological processes (12,17,18). Recently, immortalized
chondrocytes have been developed to overcome this problem (14,19). Several
immortalized chondrocyte cell lines have been shown to express a chondrocyte-
specific phenotype with a high proliferation rate. However, phenotypic stability
is lost more quickly during expansion in serial monolayer cultures, compared
with primary cultures of cells from juvenile animals.
Finally, many genetically modified animal models have been developed
recently, providing new powerful tools for understanding cartilage differen-
tiation or degradation. However, attempts to culture mouse articular
chondrocytes have been frustrated by the small size of the joints, which limits
the feasibility of extracting chondrocytes from the cartilage matrix. In addi-
tion to the method for culturing chondrocytes from numerous species, a new
method for culturing articular chondrocytes from newborn mice is detailed.
Fig. 1. Cartilage digestion chamber. Bottle used for isolating chondrocytes. A nylon
mesh forms a barrier between the inner and outer chambers. The filter is fastened to a
glass tube with a glass ring (not shown here) and placed on a glass-rod triangle, in
order to leave a narrow space between the filter and the bottom of the vessel. The
chondrocytes released into this space are withdrawn with a pipet. A magnetic bar is
placed in the inner chamber. A, glass cylinder; B, inner chamber; C, cartilage pieces;
D, magnetic bar; E, outer chamber; F, nylon mesh (48-µm pore size); G, platform; H,
battery-powered magnetic stirrer.
2. Materials
2.1. Human Primary Articular Chondrocytes
1. Cell culture medium: Dulbecco’s modified Eagle’s medium (DMEM high glu-
cose, Sigma, France), supplemented with 4 mM L-glutamine, 100 U/mL penicil-
lin, and 0.1 mg/mL streptomycin.
2. Complete culture medium consists of DMEM high glucose supplemented with
antibiotics as above and with 10% (v/v) fetal calf serum (FCS). Store FCS aliquots
at –20°C and do not refreeze after thawing (see Note 1).
3. Sterile Dulbecco’s phosphate-buffered Ca2+- and Mg2+-free saline (PBS).
4. Enzymes (Roche Diagnostics, Meylan, France): hyaluronidase 0.05% (w/v) in
PBS; trypsin, 0.2% (w/v) in PBS; collagenase A from Clostridium histolyticum,
0.2% (w/v) in PBS (see Note 2). Enzyme solutions are freshly prepared and fil-
tered through a sterile 0.22-µm filter.
5. Scalpels.
6. 30-mL Flat-bottomed glass vial, glass ring, glass cylinder, and glass triangle (cus-
tom glassware prepared by a local glassblower and adapted as shown in Fig. 1).
7. Magnetic stirrer (battery-powered).
8. Nylon mesh, 48 µm (WWR for Sefar AG, Rüschlikon, Switzerland).
9. Tissue culture plastic: sterile pipets, culture flasks, and Petri dishes.
10. CO2 incubator.
11. Centrifuge.
12. Inverted light microscope.
13. Hemocytometer.
2.2. Isolation and Culture of Rabbit

Primary Articular Chondrocytes
The following was adapted from ref. 20.
1. Cell culture medium: Ham’s F-12 medium (Sigma, France) supplemented with
4 mM L-glutamine, 100 U/mL penicillin, and 0.1 mg/mL streptomycin.
2. Complete culture medium consists of Ham’s F-12 medium supplemented with
antibiotics as above and with 10% (v/v) FCS.
3. Animal shaver.
4. 15-mL Custom glassware as shown in Fig. 1.
5. All other materials and equipment as in Subheading 2.1. (For enzymes, see
Note 3.)
2.3. Isolation and Culture of Rat

and Bovine Primary Articular Chondrocytes
1. Cell culture medium: DMEM/Ham’s F-12 medium supplemented with 2 mM
L-glutamine, 50 µg/mL gentamicin, and 0.25 µg/mL amphotericin B.
2. Complete culture medium consists of DMEM/Ham’s F-12 medium supplemented
as above with, in addition, 10% (v/v) FCS.
3. Enzymes (Roche Diagnostics): pronase 1% (w/v) in cell culture medium contain-
ing 5% FCS; bacterial collagenase 0.4% (w/v) in cell culture medium containing
5% FCS. Enzyme solutions are freshly prepared and filtered through a sterile
0.22-µm filter.
4. 15-mL Custom glassware as shown in Fig. 1.
5. All other materials and equipments as in Subheading 2.1.
2.4. Isolation and Culture of Newborn Mouse Rib Chondrocytes

1. Cell culture medium: DMEM (Sigma, France) supplemented with 2 mM
L-glutamine, 50 U/mL penicillin, and 0.05 mg/mL streptomycin.
2. Complete culture medium consists of DMEM supplemented as above with, in
addition, 10% (v/v) FCS.
3. Enzyme (Roche Diagnostics): collagenase D from Clostridium histolyticum, 3 mg/
mL in cell culture medium (see Note 2). The enzyme solution is freshly prepared
and filtered through a sterile 0.22-µm filter.
4. 70% Ethanol.
5. All other materials and equipment as in Subheading 2.1.
2.5. Isolation and Culture of Mouse Primary Articular Chondrocytes

1. All solutions, materials, and equipment as in Subheading 2.4. plus 0.5 mg/mL
collagenase D in cell culture medium.
3. Methods
3.1. Human Primary Articular Chondrocytes (see Note 4)
3.1.1. Collection of Articular Specimens
As soon as possible after surgery, place the knees or hips in cold collection
medium and store at 4°C. Specimens collected and stored in this way can be
used for chondrocyte isolation up to 48 h after surgery.
Osteoarthritic cartilage specimens can be obtained from the tibial plateaus and
femoral condyles of adults undergoing total knee replacement. They are excised
from the superficial and middle layers, avoiding the calcified layer. Nonarthritic
human articular cartilage can be isolated from femoral heads of patients with
displaced femoral neck fractures or at autopsy of individuals with no history of
joint disease and with normal cartilage by gross examination and microscopy.
Each culture is run with chondrocytes from a single patient (see Note 5).
3.1.2. Isolation and Culture of Human Chondrocytes
1. Under a laminar flow hood, place the joints in a dish containing complete culture
medium. When necessary, remove mesenchymal repair tissue with scissors and
scalpels to clear the cartilaginous layer completely. Cut across the surface to
obtain full-thickness strips of cartilage, excluding subchondral bone, and tip the
strips of cartilage into 10-cm dishes containing 0.05% hyaluronidase in PBS.
2. Finely mince all collected cartilage slices into about 1–3-mm3 pieces, using
scalpels.
3. Remove the hyaluronidase solution and rinse the cartilage once with PBS.
4. Transfer the pieces of cartilage from the Petri dish to the inner chamber of the
sterile mounted glass vial (Fig. 1) and carefully add to the inner chamber 10 mL
of 0.2% trypsin solution.
5. Transfer the vial to the CO2 incubator previously equipped with a battery-pow-
ered shaker and subject the mixture to magnetic stirring at low speed at 37°C for
45 min.
6. Discard the trypsin solution from the outer chamber and resuspend the cartilage
slices in 10 mL of 0.2% collagenase solution.
7. Incubate the vial in a CO2 incubator at 37°C for 90 min while shaking.
8. Aspirate the suspension from the outer chamber of the vial and use a pipet to
transfer it to a new sterile 50-mL polypropylene tube.
9. Add 10 mL fresh 0.2% collagenase solution to the inner chamber and incubate
for 45 min at 37°C in a CO2 incubator while shaking.
10. During step 9, centrifuge the 50-mL polypropylene tube for 5 min at 200g and
discard the supernatant.
11. Resuspend the pellet in 10 mL of complete culture medium.
12. At the end of the second collagenase digestion, aspirate the suspension from the
outer chamber of the vial and add to the 50-mL tube containing the first sus-
pended cells (see step 11).
Fig. 2. Micrographs of different species of primary cultured chondrocytes. (A) Human

articular chondrocytes. (B) Rabbit articular chondrocytes. (C) Mouse rib chondrocytes.
(D) Mouse articular chondrocytes. Phase-contrast micrograph. Original magnification
×120; scale bars, 100 µm.
13. Wash the remaining cartilage pieces in adding 10 mL serum-free culture medium
to the inner chamber of the vial. Incubate for 90 min at 37°C in a CO2 incubator
while shaking. During this step, centrifuge the 50-mL polypropylene tube for 5
min at 200g and discard the supernatant.
14. Resuspend the pellet in 10 mL of complete culture medium.
15. Pipet the cell suspension from the outer chamber of the vial and add to the 50-mL
tube containing the suspended cells.
16. Centrifuge for 5 min at 200g and discard the supernatant.
17. Resuspend the pellet in 10 mL of complete culture medium and count the cells in
a hemocytometer. Bring up to volume with complete culture medium and seed
the suspended chondrocytes onto tissue culture plates or dishes at a density of 1 ×
105 cells/cm2.
18. Incubate the culture plates or dishes for 2 d undisturbed at 37°C under humidified
conditions and 5% CO2/95% air to ensure strong attachment.
19. Change the medium on the day before harvest. To avoid dedifferentiation, we use
human primary chondrocytes between 3 and 6 d after seeding. (Their morpho-
logical appearance is shown in Fig. 2A.)
3.2. Isolation and Culture of Rabbit Primary Articular Chondrocytes

1. Sacrifice 3-wk-old female Fauve-de-Bourgogne rabbits under general anesthesia
(see Note 6). Shave each limb of the animal and clean with a disinfectant; use
sterile equipment to remove the joints.
2. Dissected out the shoulders, knees, and femoral heads of the rabbit and transfer
to 50-mL sterile polypropylene tubes containing complete culture medium.
3. Under a laminar flow hood, place the joints in a dish containing complete culture
medium. Remove extraneous tissue with scissors and scalpels to clear the carti-
laginous layer completely.
4. Cut the cartilage into longitudinal slices, and tip the slices into 10-cm dishes
containing 0.05% hyaluronidase in PBS.
5. Repeat steps 2–15 of Subheading 3.1.2., but add only 5 mL of enzyme in steps
4, 6, and 9 and 5 mL of complete culture medium in steps 11–17. Incubate for 60
min in steps 7 and 13 and for 30 min in step 9. Use a 15-mL polypropylene tube
instead of a 50-mL tube.
6. Transfer the cell suspension to a new sterile 50-mL polypropylene tube, centri-
fuge for 5 min at 200g, and discard the supernatant.
7. Resuspend the pellet in 20 mL complete culture medium and count the cells in a
hemocytometer. Bring up to volume with complete culture medium and seed the
suspended chondrocytes onto tissue culture plates or dishes at a density of 8 ×
103 cells/cm2 (see Note 7).
9. After 4 d, replace the medium with fresh complete culture medium every 2–3 d
and monitor the progress of the cultures by observing them under a phase-con-
trast inverted microscope. The chondrocytes expand and may grow to confluence
within 6–8 d (see Fig. 2B).
3.3. Isolation and Culture of Rat

and Bovine Primary Articular Chondrocytes
3.3.1. Tissue Samples
1. Bovine.
a. Harvest bovine articular cartilage from various joints of young calves (about
2–4 wk old) or adult steers (18–24 mo old), obtained from the local slaughter-
house.
b. Depending on the studies, aseptically collect full-thickness articular cartilage
slices from the metatarsophalangeal (23), metacarpophalangeal (21),
humeroscapular (24), or femoropatellar (22) joints.
c. Immediately place the articular cartilage slices in culture medium.
2. Rats
a. Male Wistar rats (130–150 g) are housed under controlled temperature and
lighting conditions, with food and water ad libitum.
b. Sacrifice the rats under general anesthesia (ketamine and acepromazine), and
cut into slices joint cartilage taken surgically from the knees and hips (see
Note 6).
3.3.2. Isolation and Culture of Chondrocytes

1. Tip slices of cartilage into 10-cm dishes containing 1% pronase (w/v) in culture
medium containing 5% FCS (see Note 8).
2. Finely mince all collected cartilage slices into about 1-mm3 pieces, using scalpels.
3. Incubate the cartilage fragments in a CO2 incubator at 37°C, for 90 min.
4. Remove the pronase solution and rinse the cartilage once with PBS.
5. Transfer the pieces of cartilage from the Petri dish to the inner chamber of the
digestion sterile glass vial and carefully add 5 mL of 0.4% collagenase solution
to the inner chamber (see Note 9).
6. Transfer the vial to the CO2 incubator previously equipped with a battery-pow-
ered shaker and subject the mixture to magnetic stirring at low speed at 37°C for
180 min.
7. Aspirate the suspension from the outer chamber of the vial and use a pipet to
transfer it to a new sterile 15-mL polypropylene tube.
8. Centrifuge the 15-mL polypropylene tube for 5 min at 200g and discard the
supernatant.
9. Resuspend the pellet in 10 mL PBS.
10. Centrifuge the 15-mL polypropylene tube for 5 min at 200g and discard the
supernatant.
11. Repeat steps 9 and 10.
12. Resuspend the pellet in 5 mL complete culture medium.
13. Count the cells in a hemocytometer. Bring up to volume with complete culture
medium and seed the suspended chondrocytes onto tissue culture plates or dishes
at a density of 1.5 × 105 cells/cm2.
15. After 4 d, replace the medium with fresh complete culture medium every 2–3 d
and monitor the progress of the cultures by observing them under a phase-con-
trast inverted microscope. The chondrocytes expand and may grow to confluence
within 6–8 d.
3.4. Isolation and Culture of Newborn Mouse Rib Chondrocytes

1. Sacrifice newborn mice (0–4 d old) under general anesthesia (see Note 6).
2. After washing with 70% ethanol, place the mouse on its back and keep in posi-
tion by pinning the mouse to a paper-covered corkplate.
3. Make a longitudinal incision through the skin of the abdomen and rib cage, pull
the skin to the sides, and open the front of the rib cage longitudinally along the
midline.
4. Dissect the ventral part of the rib cage and place in a sterile 50-mL polypropylene
tube containing cold PBS.
5. Under a laminar flow hood, wash the rib cages two or three times with sterile
PBS.
6. Incubate the rib cages in collagenase D (about 1 mL/rib cage) in the CO2 incuba-
tor at 37°C for 90 min.
7. Check that all the soft tissues are digested or can be detached from the cartilages
with a few pipetings. If not, incubate longer.
8. Transfer the pieces into a 50-mL sterile polypropylene tube.
9. Fill with PBS; allow the rib cartilages (and bones) to settle for just a few minutes
and immediately remove the supernatant.
10. Fill again with PBS, mix gently, allow the cartilages to settle, and remove the
supernatant.
11. Repeat steps 9 and 10 until the cartilages seem clean. These washes must elimi-
nate most of the soft tissues contaminating the cartilages.
12. Resuspend the cartilages in collagenase D (about 1 mL/cage) and transfer to a
10-cm Petri dish. (Do not leave in the tube, as this would cause the cells to die.)
13. Leave in the incubator for 5–6 h or until the cartilages are completely digested.
(Withdraw and expel in the pipet a few times to check digestion.) At the end of
the incubation, only the bony parts of the ribs are left undigested (see Note 10).
14. Transfer the supernatant, which contains the chondrocytes, to a new 50-mL ster-
ile polypropylene tube (avoid all bony particles).
15. Fill the tube with PBS
16. Centrifuge at 200g for 10 min to pellet the cells.
17. Remove the supernatant.
18. Repeat steps 15–17.
19. Fill the tube with complete medium.
20. Centrifuge at 200g for 10 min to pellet the cells.
21. Remove the supernatant.
22. Resuspend the cells in 8 mL of complete medium and count them.
23. Plate the chondrocytes at high density, 1–2 million cells per 10-cm2 dish in 8 mL
of complete medium (see Note 11 and Fig. 2C).
3.5. Isolation and Culture of Mouse Primary Articular Chondrocytes

1. Sacrifice newborn mice (5 d old) under general anesthesia (see Notes 6 and 12)
2. After washing with 70% ethanol, place the mouse on its abdomen and keep in
position by pinning the mouse to a paper-covered corkplate.
3. Remove the skin and soft tissue from the hind legs using scissors and one pair of
sterile curved forceps.
4. Dislocate the femur from the hip using scissors and remove the soft tissues sur-
rounding the femoral head.
5. Block the femoral head (which resembles the head of a pin) between the curved
tips of forceps, and clamp hard to cut it off. Pick up the femoral head carefully
and place it in a 50-mL sterile polypropylene tube containing complete culture
medium.
6. Incise the knee joint capsules to expose the femoral condyles and the tibial plateau.
7. Crack the bone open to isolate the femoral condyles and the tibial plateau and
place in the 50-mL tube containing the femoral heads.
8. Transfer the pieces of tissue from the tube to a sterile 100-mm Petri dish contain-
ing 12 mL collagenase D, 3 mg/mL, in cell culture medium.
9. Incubate for 45 min at 37°C in the incubator.
10. Use a sterile plastic 25-mL pipet to agitate the tissue fragments for 30 s and
transfer the pieces to a new sterile 100-mm Petri dish containing fresh collage-
nase D solution (3 mg/mL).
11. Incubate typically for 45 min at 37°C in the incubator and check that all the soft
tissues have then been removed.
12. Use a sterile plastic 25-mL pipet to agitate the tissue fragments for 30 s and
transfer the pieces to a new sterile 100-mm Petri dish containing 20 mL collage-
nase D, 0.5 mg/mL, overnight at 37°C.
13. Dislodge the smaller sheets of cells from the bone heads by vigorous agitation, first
with a sterile plastic 5-mL pipet and then with a plastic 2-mL pipet. The cell sus-
pension should be well mixed to disperse any cell aggregates in order to obtain a
suspension of single cells.
14. Allow the Petri dish to stand for 2 min to let the bone fragments to settle to the
bottom of the dish. Then draw off the cell suspension.
15. Pass the cell suspension through sterile 48-µm nylon mesh into a fresh 50-mL
polypropylene tube in order to remove any sheets of dead cells.
16. Centrifuge for 10 min at 200g and discard the supernatant.
17. Resuspend the pellet in 10 mL complete culture medium and count the cells in a
hemocytometer. Bring up to volume with complete culture medium and seed the
suspended chondrocytes onto tissue culture plates or dishes at a density of 8 × 103
cells/cm2. A 6-d-old culture is shown in Fig. 2D.
4. Notes
1. Batches of serum vary in their ability to support the proliferation and differenti-
ated phenotype expression of chondrocytes in primary culture. It is advisable to
screen the batches and to keep a large reserve of serum once a suitable batch has
been identified. We used serum from Gibco-BRL (Cergy Pontoise, France).
2. The collagenase most commonly used for tissue dissociation is a crude prepara-
tion from C. histolyticum containing clostridiopeptidase A in addition to a num-
ber of other proteases, polysaccharidases, and lipases. Crude collagenase is
apparently ideal for tissue dissociation since it contains the enzyme required to
attack native collagen, in addition to the enzymes that hydrolyze the other pro-
teins, polysaccharides, and lipids in the extracellular matrix of tissues. Collage-
nases A, B, and D are prepared from extracellular C. histolyticum culture filtrate.
These crude preparations contain collagenase and other proteases, including
clostripain, trypsin-like activity, and a neutral protease. This mixture of enzyme
activities makes crude collagenases ideal for gentle tissue dissociation to gener-
ate single cells. Collagenases A, B, and D contain different ratios of the various
proteolytic activities. This allows for selection of the preparation best suited for
disaggregation of a particular tissue. In our hands, the best results were obtained
using collagenase A for rabbit and human chondrocyte dissociation and collage-
nase D for mouse chondrocyte dissociation.
3. An alternative method for rabbit articular chondrocyte dissociation uses pronase
E (2 mg/g cartilage, 30 min, 37°C) instead of trypsin (25).
4. Many protocols have been described for isolating chondrocytes from adult human
joints (26–28). Some studies have used fetal human chondrocytes (29,30), and
in general the fetal human joint has been found to be superior to the adult human
joint in terms of cell numbers. However, the qualitative differences between fetal
and adult cells cannot be disregarded, as fetal tissues consist mainly of epiphyseal
cartilage. Based on the various published modifications of isolation techniques,
we describe here a protocol that, in our laboratory, provides both high yields and
good preservation of phenotypic properties.
5. Some authors divided the cartilage specimens of each donor into more severely
degraded and less severely degraded samples (31).
6. Ethical guidelines for experimental investigations in animals must be followed.
In particular, the experimental procedure for euthanasia must be discussed and
accepted by the local ethics committee for animal experimentation.
7. To ensure plating at a uniform density, repeatedly aspirate and expel the cell
suspension during its distribution among the dishes. After completion of the full
digestion procedure, this usually provides 20 × 106 cells per rabbit.
8. A method that has been used successfully for isolating bovine chondrocytes is
digestion with trypsin 0.25% in Hanks’ solution for 20 min followed by collage-
nase 0.25% in culture medium for 40 min (32) . This protocol has also been
adapted for rat chondrocyte isolation (33), with minor changes such as a longer
collagenase incubation (180 min) (34).
9. A commonly used alternative involves digestion with 1.5% (w/v) bacterial colla-
genase B in culture medium without serum, overnight at 37°C (35,36).
10. Alternatively, digestion can be performed overnight by diluting the collagenase
six-fold in complete culture medium.
11. In our hands, one mouse yields about 2 million chondrocytes.
12. Younger mouse pups have tiny joints that are very difficult to handle and easy to
tear. Older mice provide fewer chondrocytes per joint, probably because their cells
divide more slowly. The instruments used are autoclaved before the procedure. An
entire litter of mouse pups is usually handled at one sitting, and the instruments are
resterilized between pups by dipping in 70% ethanol before each use.
Acknowledgments
We are grateful to Colette Salvat for her outstanding technical expertise in
the development and optimization of cultured mouse articular chondrocytes and
to Lydie Humbert and Audrey Pigenet for their high-quality technical assis-
tance. We thank Claire Jacques for her expertise and detailed information about
cultured mouse rib chondrocytes. We are indebted to Professor C. Sautet (Saint
Antoine UFR, Paris) for providing human articular cartilage.
References
1.
1 Aydelotte, M. B., and Kuettner, K. E. (1988) Differences between sub-popula-
tions of cultured bovine articular chondrocytes. I. Morphology and cartilage
matrix production. Connect. Tissue Res. 18, 205–222
2.
2 von der Mark, K. and Conrad, G. (1979) Cartilage cell differentiation. Clin.
Orthop. 139, 185–205
3.
3 Stokes, D. G., Liu, G., Coimbra, I. B., Piera-Velazquez, S., Crowl, R. M., and
Jimenez, S. A. (2002) Assessment of the gene expression profile of differentiated
and dedifferentiated human fetal chondrocytes by microarray analysis. Arthritis
Rheum. 46, 404–419.
4.
4 Kergosien, N., Sautier, J., and Forest, N. (1998) Gene and protein expression dur-
ing differentiation and matrix mineralization in a chondrocyte cell culture system.
Calcif. Tissue Int. 62, 114–121.
5.
5 Hauselmann, H. J., Aydelotte, M. B., Schumacher, B. L., Kuettner, K. E., Gitelis,
S. H., and Thonar, E. J. (1992) Synthesis and turnover of proteoglycans by human
and bovine adult articular chondrocytes cultured in alginate beads. Matrix 12,
116–129.
6.
6 Karsenty, G. (2001) Chondrogenesis just ain’t what it used to be. J. Clin. Invest.
107, 405–407.
7.
7 Corvol, M. T., Dumontier, M. F., and Rappaport, R. (1975) Culture of
chondrocytes from the proliferative zone of epiphyseal growth plate cartilage from
prepubertal rabbits. Biomedicine 23, 103–107.
8.
8 Mwale, F., Billinghurst, C., Wu, W., et al. (2000) Selective assembly and remod-
elling of collagens II and IX associated with expression of the chondrocyte hyper-
trophic phenotype. Dev. Dyn. 218, 648–662.
9.
9 Goldring, M. B. (2000) The role of the chondrocyte in osteoarthritis. Arthritis
Rheum. 43, 1916–1926.
10.
10 Goldring, M. B. and Berenbaum, F. (1999) Human chondrocyte culture models
for studying cyclooxygenase expression and prostaglandin regulation of collagen
gene expression. Osteoarthritis Cartilage 7, 386–388.
11.
11 Watt, F. M. (1988) Effect of seeding density on stability of the differentiated phe-
notype of pig articular chondrocytes in culture. J. Cell. Sci. 89, 373–378.
12.
12 Stokes, D. G., Liu, G., Dharmavaram, R., Hawkins, D., Piera-Velazquez, S.,
and Jimenez, S. A. (2001) Regulation of type-II collagen gene expression dur-
ing human chondrocyte de-differentiation and recovery of chondrocyte-specific
phenotype in culture involves Sry-type high-mobility-group box (SOX) tran-
scription factors. Biochem. J. 360, 461–470.
13.
13 Goldring, M. B., Birkhead, J., Sandell, L. J., Kimura, T., and Krane, S. M. (1988)
Interleukin 1 suppresses expression of cartilage-specific types II and IX collagens
and increases types I and III collagens in human chondrocytes. J. Clin. Invest. 82,
2026–2037.
14.
14 Goldring, M. B., Birkhead, J. R., Suen, L. F., et al. (1994) Interleukin-1 beta-
modulated gene expression in immortalized human chondrocytes. J. Clin. Invest.
94, 2307–2316.
15.
15 Demoor-Fossard, M., Redini, F., Boittin, M., and Pujol, J. P. (1998) Expression
of decorin and biglycan by rabbit articular chondrocytes. Effects of cytokines and
phenotypic modulation. Biochim. Biophys. Acta 1398, 179–191.
16.
16 Gibson, G. J., Schor, S. L., and Grant, M. E. (1982) Effects of matrix macro-
molecules on chondrocyte gene expression: synthesis of a low molecular weight
collagen species by cells cultured within collagen gels. J. Cell Biol. 93, 767–
774.
17.
17 Takigawa, M., Pan, H. O., Kinoshita, A., Tajima, K., and Takano, Y. (1991)
Establishment from a human chondrosarcoma of a new immortal cell line with
high tumorigenicity in vivo, which is able to form proteoglycan- rich cartilage-
like nodules and to respond to insulin in vitro. Int. J. Cancer 48, 717–725.
18.
18 Mallein-Gerin, F., and Olsen, B. R. (1993) Expression of simian virus 40 large
T (tumor) oncogene in mouse chondrocytes induces cell proliferation without
loss of the differentiated phenotype. Proc. Natl. Acad. Sci. USA 90, 3289–3293.
19.
19 Robbins, J. R., Thomas, B., Tan, L., et al. (2000) Immortalized human adult articu-
lar chondrocytes maintain cartilage- specific phenotype and responses to
interleukin-1β. Arthritis Rheum. 43, 2189–2201.
20. Adolphe, M., Froger, B., Ronot, X., Corvol, M. T., and Forest, N. (1984) Cell
multiplication and type II collagen production by rabbit articular chondrocytes
cultivated in a defined medium. Exp. Cell Res. 155, 527–536.
22. Martin, I., Vunjak-Novakovic, G., Yang, J., Langer, R., and Freed, L. E. (1999)
Mammalian chondrocytes expanded in the presence of fibroblast growth factor 2
maintain the ability to differentiate and regenerate three-dimensional cartilagi-
nous tissue. Exp. Cell Res. 253, 681–688.
21. Kuettner, K. E., Memoli, V. A., Pauli, B. U., Wrobel, N. C., Thonar, E. J., and Daniel,
J. C. (1982) Synthesis of cartilage matrix by mammalian chondrocytes in vitro. II.
Maintenance of collagen and proteoglycan phenotype. J. Cell Biol. 93, 751–757.
23. Domm, C., Schunke, M., Christesen, K., and Kurz, B. (2002) Redifferentiation of
dedifferentiated bovine articular chondrocytes in alginate culture under low oxy-
gen tension. Osteoarthritis Cartilage 10, 13–22.
24. Zaucke, F., Dinser, R., Maurer, P., and Paulsson, M. (2001) Cartilage oligomeric
matrix protein (COMP) and collagen IX are sensitive markers for the differentia-
tion state of articular primary chondrocytes. Biochem. J. 358, 17–24.
25. Rahfoth, B., Weisser, J., Sternkopf, F., Aigner, T., von der Mark, K., and Brauer,
R. (1998) Transplantation of allograft chondrocytes embedded in agarose gel into
cartilage defects of rabbits. Osteoarthritis Cartilage 6, 50–65.
26. Robbins, J. R. and Goldring, M. B. (1998) Preparation of immortalized human
chondrocyte cell lines, in Tissue Engineering, Vol. 18 (Morgan, J. R., and
Yarmush, M. L., eds.), Humana, Totowa, NJ, pp. 173–192.
27. Goldring, M. B. (1996) Human chondrocyte cultures as models of cartilage-spe-
cific gene regulation, in Human Cell Culture Protocols, Vol. 2 (Gareth, E. J., ed.),
Humana, Totowa, NJ, pp. 217–232.
28. Aulthouse, A. L., Beck, M., Griffey, E., et al. (1989) Expression of the human
chondrocyte phenotype in vitro. In Vitro Cell Dev. Biol. 25, 659–668.
29. Carrascosa, A., Audi, L., and Ballabriga, A. (1985) Morphologic and metabolic
development of human fetal epiphyseal chondrocytes in primary culture. Pediatr.
Res. 19, 720–727.
30. Reginato, A. M., Iozzo, R. V., and Jimenez, S. A. (1994) Formation of nodular
structures resembling mature articular cartilage in long-term primary cultures of
human fetal epiphyseal chondrocytes on a hydrogel substrate. Arthritis Rheum.
37, 1338–1349.
31. Stove, J., Gerlach, C., Huch, K., et al. (2001) Gene expression of stromelysin and
aggrecan in osteoarthritic cartilage. Pathobiology 69, 333–338.
32. Kawiak, J., Moskalewski, S., and Darzynkiewicz, Z. (1965) Isolation of
chondrocytes from calf cartilage. Exp. Cell Res. 39, 59–68.
33. Nedelec, E., Abid, A., Cipolletta, C., et al. (2001) Stimulation of cyclooxygenase-
2-activity by nitric oxide-derived species in rat chondrocyte: lack of contribution
to loss of cartilage anabolism. Biochem. Pharmacol. 61, 965–978.
34. Okazaki, M., Higuchi, Y., and Kitamura, H. (2003) AG-041R stimulates cartilage
matrix synthesis without promoting terminal differentiation in rat articular
chondrocytes. Osteoarthritis Cartilage 11, 122–132.
35. Gouze, J. N., Bordji, K., Gulberti, S., et al. (2001) Interleukin-1beta down-regulates
the expression of glucuronosyltransferase I, a key enzyme priming glycosaminogly-
can biosynthesis: influence of glucosamine on interleukin-1β-mediated effects in
rat chondrocytes. Arthritis Rheum. 44, 351–360.
36. Gouze, J. N., Bianchi, A., Becuwe, P., et al. (2002) Glucosamine modulates IL-1-
induced activation of rat chondrocytes at a receptor level, and by inhibiting the
NF-k B pathway. FEBS Lett. 510, 166–170.
Culture of Chondrocytes in Alginate 15
2
Culture of Chondrocytes in Alginate Beads
Frédéric De Ceuninck, Christophe Lesur,

Philippe Pastoureau, Audrey Caliez, and Massimo Sabatini
Summary
A classic method for the encapsulation and culture of chondrocytes in alginate beads is
described. Chondrocytes are released from cartilage matrix by collagenase/dispase digestion
and mixed with a solution of 1.25% alginic acid until a homogenous suspension is obtained.
The suspension is drawn into a syringe and pushed gently through a needle, so that drops fall
into a solution of calcium chloride. Beads form instantaneously and further polymerize after
5 min in the calcium chloride solution. Chondrocytes from any species, including human
osteoarthritic chondrocytes, can be cultured with this technique. Under these conditions,
chondrocytes maintain a high degree of differentiation. Beads can be dissolved by chelation
of calcium with EDTA. In this way, chondrocytes can be recovered and further separated from
the matrix by centrifugation. Almost all molecular and biochemical techniques, as well as a
number of biological assays, are compatible with the culture of chondrocytes in alginate.
Key Words: Chondrocyte; cartilage; alginate; differentiation; cell culture.
1. Introduction
After being released from their cartilaginous matrix by enzyme digestion,
chondrocytes have a tendency to dedifferentiate, especially if they are cul-
tured at a low density in monolayer culture. The round cells rapidly lose their
cartilage phenotype and transform into flattened fibroblast-like cells. Under
these conditions, certain specific markers of the chondrocytic phenotype are
downregulated, and some nonchondrocytic proteins such as type I collagen
are synthesized (1). To prevent dedifferentation, culture at a high density
(more than 5 × 104 cells/cm2) is recommended and passage culture must be
avoided. These are important constraints, especially when working on human
osteoarthritic chondrocytes, which are often obtained in low amounts from a
single cartilage specimen.
15
16 De Ceuninck et al.
To maintain chondrocytes in their original phenotype, various methods have

been developed, including culture in scaffolding materials, such as collagen
sponges, agarose, or alginate. Among these techniques, culture in alginate is
often used, as it offers many advantages over the others. Chondrocytes can be
homogeneously dispersed and encapsulated in alginate beads and cultured for
more than 8 mo without any apparent phenotype loss (2). Alginate beads allow
the reconstitution of a tridimensional scaffold closely resembling that of the car-
tilage matrix (3). Chondrocytes express proteoglycans, especially aggrecan, and
also type II, but not type I collagen (2–8). Encapsulated chondrocytes are still
able to respond to growth factors and cytokines usually known to affect chon-
drocyte metabolism (9–11). Beads are easy to handle, and many molecular, bio-
chemical, and biological applications are compatible with this mode of culture
(2–18). Cells can be recovered from beads by simple chelation of divalent ions
with ethylenediamine tetraacetic acid (EDTA) followed by centrifugation. Cul-
ture in alginate beads is also useful to redifferentiate chondrocytes that have
dedifferentiated because of expansion in bidimensional culture (16–18). The
original method of culture of chondrocytes in alginate beads was developed in
the late eighties by Guo et al. (19) and was then employed and improved by
many researchers. We describe here the method routinely used in our laboratory.
2. Materials
1. Articular cartilage explants from guinea pigs, rabbits, or rats or from human nor-
mal or osteoarthritic cartilage.
2. Hanks’ balanced salt solution (HBSS; Gibco-BRL)
3. Ham’s F-12 culture medium with Glutamax (Ham F-12; Gibco-BRL).
4. Fetal calf serum (FCS).
5. 10,000 U/mL Penicillin /10,000 µg/mL streptomycin stock solution (PS;
Gibco-BRL).
6. Dispase/collagenase solution for guinea pig, rabbit, or rat cartilage: 2 mg/mL
dispase (from Bacillus polymixa; Gibco-BRL) and 3 mg/mL collagenase type I
(Worthington) in HBSS. Sterilize through a 0.22 µm filter.
7. Dispase/collagenase solution for human cartilage: 2 mg/mL dispase (from B. poly-
mixa; Gibco-BRL) and 3 mg/mL collagenase type I (Worthington) in Ham F-12
supplemented with 10% FCS. Sterilize through a 0.22 µm filter.
8. Cell culture equipment, including 10-mm-diameter Petri dishes, pipets,
multipipets, and appropriate tips or combitips.
9. 50-mL Capacity beaker sterilized by autoclaving.
10. Cell strainer, 40 µm nylon (Falcon, cat. no. 2340).
11. Blue max 50-mL sterile tubes (Falcon, cat. no. 2070).
12. 21-Gage needle (Terumo, cat. no. NN-2125R) with syringe of 2, 5, or 10 mL
(Terumo).
13. Sterile 0.9% NaCl solution.
14. Alginate solution: 1.25% alginic acid (Fluka, cat. no. 71238), 20 mM HEPES,
150 mM NaCl, pH 7.4. First dissolve HEPES and NaCl powders in deionized
water. Warm up the mixture at 60°C and add the alginate powder with constant
stirring until the solution is homogeneous. It may take more than 1 h to achieve
complete dissolution. Let the solution cool down to room temperature and adjust
to pH 7.4 (see Note 1) . Adjust the final volume of the solution with deionized
water. Autoclave.
15. Polymerization solution: 102 mM CaCl2, 10 mM HEPES, pH 7.4. Pass through a
0.22 µm-filter.
16. Dissolution solution: 55 mM EDTA, 10 mM HEPES, pH 7.4. Pass through a
0.22 µm-filter.
17. Tissue culture incubator, set at 37°C, with water-saturated, 5% CO2 air.
3. Methods
3.1. Chondrocyte Isolation
Chondrocytes from any species may be used with this technique.
1. Guinea pig, rat, or rabbit cartilage explants: finely mince articular cartilage down
to fragments of around 1 mm3. Transfer the fragments (0.25–0.5 g) into a Petri
dish containing 20 mL of dispase/collagenase solution in HBSS. Incubate at 37°C
for 5 h, or until the explants are digested.
2. Human cartilage: finely mince articular cartilage down to fragments of around 1
mm3 . Transfer the fragments (0.25–0.5 g) into a Petri dish containing 20 mL of
dispase/collagenase solution in Ham F-12 with 10% FCS. Incubate at 37°C for
16 h or until the explants are digested (see Notes 2 and 3).
3. Pellet cells by 3-min centrifugation at 900g.
4. Resuspend cell pellet in Ham F-12 medium supplemented with 10% FCS and 1%
PS. Count the cells on a hemacytometer, and centrifuge the suspension for 3 min
at 900g. Discard the supernatant and keep the cell pellet.
3.2. Encapsulation of Chondrocytes in Alginate Beads

The overall method is depicted in Fig. 1 and described here in detail. All
steps should be performed under a sterile hood.
1. Resuspend the cell pellet obtained in Subheading 3.1., step 4 by adding a vol-
ume of alginate solution (Fig. 1A), so as to have 2 million cells/mL of alginate
solution (see Note 4).
2. Aspirate the suspension into a syringe and cap with a 21-gage needle.
3. Gently push the piston of the syringe so that the solution is released dropwise
into 30 mL of polymerization solution, maintained under gentle stirring by a
magnet, in a 50 mL sterile beaker (Fig. 1B).
4. Beads instantly polymerize when falling into the solution, entrapping the
chondrocytes. Let beads polymerize completely under gentle agitation for an ad-
ditional 10 min.
18
Fig 2. (A) Morphological appearance of alginate beads with encapsulated rabbit

chondrocytes. Beads were cultured in medium containing 10% FCS. The increase of
cell density was evident at d 12. Cell density was further increased at d 26. (B) Macro-
scopic view of beads as a function of time. The volume of beads increased by about
40% at d 26.
5. Pour the solution containing the beads on the cell strainer laid on top of a 50-mL
Falcon tube. Discard the filtered polymerization solution and carefully pick up
the beads with a spatula.
6. Transfer the beads into 30 mL of a sterile 0.9% NaCl solution (see Note 5) in a
50-mL capacity beaker and wash beads by gentle stirring for 1–2 min.
7. Repeat step 6 three times (Fig. 1C) and finally rinse beads with complete culture
medium.
8. The gross appearance of alginate beads with entrapped rabbit chondrocytes cul-
tured in FCS-containing medium is shown in Fig. 2. The granular aspect and
increase of density of beads at d 26 postencapsulation accounts for the prolifera-
tion of chondrocytes and the synthesis of matrix components. By contrast, human
osteoarthritic chondrocytes do not proliferate in alginate beads (Fig. 3). However,
they are still capable of synthetic activities.
Fig 1. (opposite page) Schematic representation of the encapsulation of chondrocytes

in alginate beads. (A) Isolated cells are suspended at 2 million/mL of 1.25% alginic
acid solution. (B) The mixture is aspirated in a syringe, and then released dropwise into
a calcium chloride solution by passage through a 21-gage needle. Beads form instantly
and completely polymerize by 10 min in this solution. (C) Beads undergo three washes
in a 0.9% sterile NaCl solution and are then ready for culture.
Fig. 3. (A) Morphological appearance of an alginate bead with encapsulated human

osteoarthritic chondrocytes 12 wk after encapsulation. Unlike the rabbit chondrocytes
in Fig. 2, human osteoarthritic chondrocytes do not proliferate in alginate beads cul-
tured in serum-containing medium. (B) Magnification (×5) of the bead shown in A.
9. Culture beads as long as required in Petri dishes in Ham F-12 medium containing
Glutamax plus 10% FCS and 1% PS. Replace medium every 2 d. In our hands,
chondrocytes need about 3 wk to reestablish a consistent pericellular and territo-
rial extracellular matrix.
3.3. Experimental Procedures: Recommendations

1. Perform experiments in serum-free medium if required, after a prior 24-h wash-
ing-out period to ensure that all interfering serum constituents have been removed
from the beads.
2. For certain experimental procedures, it may be necessary to separate cells from
the alginate matrix: for this, incubate beads in dissolution solution (at a ratio of
about 200 µL/bead) for 5 min. Centrifuge. Recover cells in the pellet and the
matrix in the supernatant.
3. Depending on the experiments, the results can be normalized relative to the weight
of beads, the number of chondrocytes, or the DNA content within each bead.
4. Notes
1. Care must be taken not to exceed the pH 7.4 limit since any attempt to bring back
pH to 7.4 with HCl may lead to precipitation.
2. Digestion of human cartilage is slower than that of cartilage from rat, rabbit, or
guinea pig, even when the same digestion volume/cartilage weight is used. To
ensure chondrocyte viability, the digestion is performed in culture medium
supplemented with FCS, even if the presence of endogenous collagenase inhibi-
tors in FCS may somewhat slow down the digestion. For reproducible digestion
conditions, it is recommended to maintain the same enzyme/tissue ratio and to
distribute the explants, if necessary, across several Petri dishes.
3. For a similar weight of explants before digestion, the number of cells obtained
from human cartilage will not be more than half that obtained from cartilage of
rat, rabbit, or guinea pig and will be highly dependent on the age of the donor and
stage of osteoarthritis.
4. Expect about 200–300 drops, i.e., 200–300 beads starting from a 5-mL homog-
enous suspension. This means that each bead will contain about 33–50 × 103
cells.
5. Using other solutions (such as phosphate-buffered saline) to wash the beads may
lead to precipitation.
References
1.
1 Benya, P. D. and Shaffer, J. D. (1982) Dedifferentiated chondrocytes reexpress the
differentiated collagen phenotype when cultured in agarose gels. Cell 30, 215–224.
2.
2 Hauselmann, H. J., Fernandes, R. J., Mok, S. S., et al. (1994) Phenotypic stability
of bovine articular chondrocytes after long-term culture in alginate beads. J. Cell
Sci. 107, 17–27.
3.
3 Hauselmann, H. J., Masuda, K., Hunziker, E. B., et al. (1996) Adult human
chondrocytes cultured in alginate form a matrix similar to native human articular
cartilage. Am. J. Physiol. 271, C742–C752.
4.
4 Tamponnet, C., Ramdi, H., Guyot, J. B., and Lievremont, M. (1992) Rabbit
articular chondrocytes in alginate gel: characterization of immobilized prepara-
tions and potential applications. Appl. Microbiol. Biotechnol. 37, 311–315.
5.
5 Hauselmann, H. J., Aydelotte, M. B., Schumacher, B. L., Kuettner, K. E.,
Gitelis, S. H., and Thonar, E., J. (1992) Synthesis and turnover of proteoglycans
by human and bovine adult articular chondrocytes cultured in alginate beads.
Matrix 12, 116–129.
6.
6 Mok, S. S., Masuda, K., Hauselmann, H. J., Aydelotte, M. B., and Thonar, E. J.
(1994) Aggrecan synthesized by mature bovine chondrocytes suspended in algi-
nate; identification of two distinct metabolic matrix pools. J. Biol. Chem. 269,
33021–33027.
7.
7 Petit, B., Masuda, K., D’Souza, A. L., et al. (1996) Characterization of crosslinked
collagens synthesized by mature articular chondrocytes cultured in alginate beads:
comparison of two distinct matrix compartments. Exp. Cell Res. 225, 151–161.
8.
8 Beekman, B., Verzijl, N., Bank, R. A., von der Mark, K., and TeKoppele, J. M.
(1997) Synthesis of collagen by bovine chondrocytes cultured in alginate; posttrans-
lational modifications and cell-matrix interaction. Exp. Cell Res. 237, 135–141.
9.
9 Redini, F., Min, W., Demoor-Fossard, M., Boittin, M., and Pujol, J. P. (1997)
Differential expression of membrane-anchored proteoglycans in rabbit articular
chondrocytes cultured in monolayers and in alginate beads. Effect of transform-
ing growth factor-beta 1. Biochim. Biophys. Acta 1355, 20–32.
10.
10 Beekman, B., Verzijl, N., de Roos, J. A., and TeKoppele, J. M. (1998) Matrix
degradation by chondrocytes cultured in alginate: IL-1β induces proteoglycan
degradation and proMMP synthesis but does not result in collagen degradation.
Osteoarthritis Cartilage 6, 330–340.
11.
11 Loeser, R. F., Todd, M. D., and Seely, B. L. (2003) Prolonged treatment of human
osteoarthritic chondrocytes with insulin-like growth factor-I stimulates
proteoglycan synthesis but not proteoglycan matrix accumulation in alginate cul-
tures. J. Rheumatol. 30, 1565–1570.
12.
12 Chubinskaya, S., Huch, K., Schulze, M., Otten, L., Aydelotte, M. B., and Cole, A.
A. (2001) Gene expression by human articular chondrocytes cultured in alginate
beads. J. Histochem. Cytochem. 49, 1211–1219.
13.
13 Stove, J., Fiedler, J., Huch, K., Gunther, K. P., Puhl, W., and Brenner, R. (2002)
Lipofection of rabbit chondrocytes and long lasting expression of a lacZ reporter
system in alginate beads. Osteoarthritis Cartilage 10, 212–217.
14.
14 Knight, M. M., van de Breevaart Bravenboer, J., Lee, D. A., van Osch, G. J.,
Weinans, H., and Bader, D. L. (2002) Cell and nucleus deformation in compressed
chondrocyte-alginate constructs: temporal changes and calculation of cell modu-
lus. Biochim. Biophys. Acta 1570, 1–8.
15.
15 Enobakhare, B. O., Bader, D. L., and Lee, D. A. (1996) Quantification of sulfated
glycosaminoglycans in chondrocyte/alginate cultures, by use of 1,9-dimethyl-
methylene blue. Anal. Biochem. 243, 189–191.
16.
16 Bonaventure, J., Kadhom, N., Cohen-Solal, L., et al. (1994) Reexpression of car-
tilage-specific genes by dedifferentiated human chondrocytes cultured in alginate
beads. Exp. Cell Res. 212, 97–104.
17.
17 Lemare, F., Steimberg, N., Le Griel, C., Demignot, S., and Adolphe, M. (1998)
Dedifferentiated chondrocytes cultured in alginate beads: restoration of the dif-
ferentiated phenotype and of the metabolic responses to interleukin-1β. J. Cell.
Physiol. 176, 303–313.
18.
18 Liu, H., Lee, Y. W., and Dean, M. F. (1998) Re-expression of differentiated
proteoglycan phenotype by dedifferentiated human chondrocytes during culture
in alginate beads. Biochim. Biophys. Acta 1425, 505–515.
19. Guo, J. F., Jourdian, G. W., and MacCallum, D. K. (1989) Culture and growth
characteristics of chondrocytes encapsulated in alginate beads. Connect. Tissue
Res. 19, 277–297.
Immortalized Chondrocyte Cell Lines 23
3
Immortalization of Human Articular Chondrocytes
for Generation of Stable, Differentiated Cell Lines
Mary B. Goldring
Summary
Immortalized chondrocytes of human origin have been developed to serve as reproducible
models for studying chondrocyte function. In this chapter, methods for immortalization of
primary human chondrocytes with SV40-TAg, HPV-16 E6/E7, and telomerase by retrovirally
mediated transduction and selection for neomycin resistance are described. However, stable
integration of an immortalizing gene stabilizes proliferative capacity, but not the differenti-
ated chondrocyte phenotype. Thus, strategies for selection of chondrocyte cell lines, involv-
ing the maintenance of high cell density and moderation of cell proliferation, are also
described. The methods for immortalization and selection are applicable to the development
of chondrocyte cell lines using any immortalizing agent. Although immortalized chondrocytes
should not be considered as substitutes for primary chondrocytes, they may be useful tools for
evaluating and further validating mechanisms relevant to cartilage biology.
Key Words: Immortalized chondrocytes; type II collagen; aggrecan; retrovirus production;

viral titer; retroviral infection; simian virus 40 large T antigen (SV40-TAg); human
papillomavirus type 16 early function genes E6 and E7 (HPV-16 E6/E7); telomerase; chondro-
cyte morphology; chondrocyte proliferation; G418; hyaluronidase; polybrene; Nutridoma.
1. Introduction
Successful development of therapeutic strategies that prevent degradation
of cartilage matrix in patients with osteoarthritis and permit cartilage regenera-
tion and repair depends on the availability of reproducible cell culture models
of human origin. Primary cultures of articular chondrocytes isolated from vari-
ous animal and human sources have served as useful models for studying the
mechanisms controlling responses to growth factors and cytokines (1). The use
of chondrocytes of human origin has been problematical, because the source of
the cartilage cannot be controlled, sufficient numbers of cells are not readily
23
24 Goldring
obtained from random operative procedures, and the phenotypic stability and
proliferative capacity in adult human chondrocytes is quickly lost upon expan-
sion in serial monolayer cultures (2). Thus, a reproducible source of
chondrocytes of human origin would be most desirable for studying cartilage
function relevant to human osteoarthritis. Furthermore, the availability of phe-
notypically stable culture systems employing immortalized human
chondrocytes would permit prior testing in vitro of strategies for improving
autologous chondrocyte transplantation, implantation of chondrocyte-laden
biomaterials, and gene transfer (3).
Several different approaches have been used in the attempt to develop cell
lines that maintain the chondrocyte phenotype. Chondrocyte cell lines that dis-
play high proliferative capacities and retain at least some features of the differ-
entiated phenotype have been derived from nonhuman sources using viral
oncogenes (4–8). Chondrocyte cell lines have also arisen spontaneously from
fetal rat calvaria (9,10) or have been derived from transgenic mice harboring
the temperature-sensitive mutant of simian virus 40 (SV40) large T antigen
(TAg) (11,12).
Recent studies have focused on adult human articular chondrocytes as target
cells for immortalization, since articular cartilage is the primary joint tissue
requiring replacement or reconstruction after it is damaged in osteoarthritis.
Stable expression of SV40-TAg using plasmid or retroviral vectors has been
used as a popular method for immortalizing human chondrocytes (13–16).
Articular chondrocyte cell lines have also been established using the human
papilloma virus type 16 (HPV-16) early function genes E6 and E7 (17) and
telomerase (18). A common finding, however, is that stable integration of immor-
talizing genes disrupts normal cell-cycle control but does not stabilize expres-
sion of the type II collagen gene (COL2A1), the most sensitive marker of the
differentiated chondrocyte phenotype. This is true even when a chondrocyte-
specific promoter is used to drive TAg expression (14). Indeed, the expression of
the differentiated phenotype in chondrocyte cell lines appears to be inversely
related to the proliferative capacity of the cell line. Thus, strategies that maintain
high cell density and decrease cell proliferation have been used to develop cell
lines that express chondrocyte-specific phenotype (15–17).
This chapter will focus on strategies for immortalization and selection that
may be applied to the development of a chondrocyte cell line using any immor-
talizing agent. Immortalized chondrocytes cannot be considered as substitutes
for primary cultures but are useful for validation and further elucidation of
mechanisms uncovered in nonimmortalized chondrocytes. Furthermore, immor-
talization of osteoarthritic chondrocytes will not necessarily stabilize character-
istics observed in primary cultures of these cells, unless they are hereditary
features of the original chondrocytes in situ. However, immortalized cell lines
may be useful as reproducible culture systems for evaluating chondrocyte func-

tions, and additional approaches are described in Chapter 4.
2. Materials
2.1. Culture Reagents
1. Growth medium for packaging cell lines: Dulbecco’s modified Eagle’s medium
(DMEM) with high glucose (4.5 g/L), L-glutamine and sodium pyruvate, and
without HEPES, containing 10% fetal calf serum (FCS).
2. Growth medium for chondrocytes: DMEM/Ham’s F-12, 1:1 (v/v). Mix equal vol-
umes of DMEM, high glucose, with L -glutamine and Ham’s F-12, with
L-glutamine (Cambrex, Walkersville, MD), both without HEPES, to give final con-
centrations of 3.151 g/L glucose, L-glutamine (365 mg/L)/L-glutamic acid (7.36
mg/L), and sodium pyruvate (110 mg/L). Add 10% FCS immediately before use.
3. Dulbecco’s phosphate-buffered Ca2+- and Mg2+-free saline (PBS).
4. Trypsin-EDTA solution: 0.05% trypsin and 0.02% ethylenediamine tetraacetic
acid (EDTA) in Hanks’ balanced salt solution without Ca2+ and Mg2+ (Life Tech-
nologies, Gaithersburg, MD).
5. Serum substitutes for experimental incubations of immortalized chondrocytes:
Nutridoma-SP (Roche Applied Science) is provided as sterile concentrate (100X,
pH 7.4; storage at 15–25°C protected from light). Dilute 1:100 (v/v) with sterile
DMEM/Ham’s F-12, without FCS, and use immediately. ITS+ (BD Biosciences)
is an alternative serum substitute.
6. Hyaluronidase (type I-S, Sigma, St. Louis, MO): Reconstituted in 50 mM sodium
acetate, pH 6.0, at 1000X stock concentration and added freshly to medium at
4 U/mL.
Except where specified, cell culture reagents can be obtained from a number
of different suppliers, including Life Technologies, Cambrex, Sigma, or
Intergen (Purchase, NY). Reserve serum testing, which is offered by these sup-
pliers, is recommended for selection of lots of FCS that maintain chondrocyte
phenotype (see Note 1). Cell culture reagents have shelf lives as recommended
by the suppliers. FCS and trypsin-EDTA are stored at –20°C but should not be
refrozen after thawing for use.
2.2. Packaging Cell Lines
The retrovirus packaging cell lines (see Note 2) that are amphotropic, pro-
ducing virus particles that infect human cells with high titer, include PA317
(ATTC, CRL-9078) (19), PT67 (BD Biosciences Clontech), or 293GPG (Phoe-
nix cell line from Gary Nolan, Stanford University) (20). They may be
obtained commercially or derived by stable transfection with the plasmid form
of the retroviral vector containing a selectable marker, such as NeoR. Replica-
tion-defective mutants of SV40-TAg, inserted in the pZipNeoSV(X) shuttle
vector, including the U19 mutant that does not bind the SV40 origin (21) and
tsA58-3 that encodes a temperature-sensitive (ts) mutant of SV40-TAg (22),
26 Goldring
may be packaged in PA317 or 293GPG. SV40-TAg-containing vectors are

generally available by request from the investigators who have developed them.
However, other immortalizing vectors may be the agents of choice. PA317-
LXSN-16E6E7 (ATTC, CRL2203) is a packaging cell line that produces HPV-
16 E6/E7. The telomerase reverse transcriptase (hTERT) expression plasmid
(C. Harley [23], Geron Corporation, Menlo Park, CA), inserted in the pLNCX
vector (BD Biosciences Clontech), is packaged in the PT67 cell line. Since the
current technology is advancing rapidly, advice should be sought from col-
laborators in the field or institutional experts.
2.3. Retrovirus Production and Infection of Chondrocytes
1. Polybrene (Sigma): 8 mg/mL in PBS (1000X stock), sterile filtered, and stored at
–20°C. Add to freshly thawed virus-containing medium to a final concentration
of 8 µg/mL.
2. G418 sulfate (neomycin analog, also known as Geneticin; Sigma or Life Tech-
nologies): 30 mg/mL in PBS (100X stock), sterile filtered, freshly prepared or
stored in aliquots at –20°C. (Avoid freeze/thawing.) Add to culture medium at a
final concentration of 300–400 µg/mL for selection and 200 µg/mL for mainte-
nance (see Note 3). For production of virus in the 293GPG packaging cell line,
also prepare 1000X stocks of puromycin (Sigma) and tetracycline (Sigma) to
give final concentrations of 2 and 1 µg/mL, respectively.
3. Antibodies for detection of immortalizing antigens, including SV40-TAg.
HPV16-E6, HPV16-E7, and hTERT are available from Santa Cruz Biotechnol-
ogy. The TRAPeze kit used to detect telomerase activity is available from
Intergen. The telomeric probe (TTAGGG)3 is from Genset (LaJolla, CA).
3. Methods
The methods describe immortalization strategies using retroviral vectors
containing NeoR as the selectable marker, although the approaches for selec-
tion and subsequent expansion are generally applicable when other types of
immortalization vectors are used. The methods for isolation and primary cul-
ture of chondrocytes and phenotypic characterization are described in Chap-
ters 1 and 14. The methods described below outline: (1) retrovirus production,
(2) retroviral infection of chondrocytes, (3) selection and establishment of
immortalized chondrocytes, and (4) characterization of established cell lines.
3.1. Retrovirus Production

3.1.1. Production of Stably Transfected Packaging Cell Line (24)
1. One day before transfection, plate packaging cells at a density of 4 × 105 per
6-cm dish or 25-cm2 flask.
2. Transfect cells 16–24 h later with 10–15 µg of plasmid DNA encoding the immor-
talizing retrovirus by the calcium-phosphate technique, using one of several com-
mercially available kits.
3. Before the cells reach confluence, 24–48 h after the transfection, split the cells
from each dish into two 10-cm dishes containing 400 µg/mL of G418 (see Note 3).
Change medium containing fresh G418 every 2–3 d. Once cell death is evident, the
G418 concentration can be changed to 200 µg/mL until G418-resistant colonies are
visible. Colonies may then be isolated, expanded, and selected for high viral titer,
or the entire G418-selected population may be expanded if the viral titer is suffi-
ciently high.
4. Liquid nitrogen storage of packaging cell line: expand the cell line in several 10-
cm dishes. When the cultures are nearly confluent, trypsinize and wash the cells in
PBS and count with a hemocytometer. Pellet the cells and resuspend at a concen-
tration of 2 × 106 cells/mL in DMEM containing 10% FCS and 10% dimethylsul-
foxide (DMSO) with gentle swirling. Pipet 1.5 mL aliquots of cell suspension into
1.8-mL cryovials. Freeze the vials at –70°C for 24 h in a styrofoam rack with a lid
or a Cryo 1°C Freezing Chamber (Nalgene). Transfer the vials to liquid nitrogen
for long-term storage. Subsequent experiments should be carried out with cultures
derived from freshly thawed cells.
3.1.2. Production of Retrovirus (24)

1. Culture the packaging cell line in DMEM containing 10% FCS at 37°C in 25-cm2
flasks. Passage the cultures twice weekly with a split ratio of 1:15 (see Note 4).
2. Collect virus-containing conditioned medium in 15-mL centrifuge tubes from
subconfluent, actively dividing cultures. Filter through a 0.45-µm filter and use
immediately, or quick-freeze on dry ice and store at –70°C in 5-mL aliquots until
needed.
3. To determine viral titer in the culture supernatants:
a. Split the producer cell line at 1:10 or 1:20 into several 6-cm dishes the day
before the infection.
b. Remove medium and add 1–2 mL of fresh medium containing different dilu-
tions of virus stock ranging from 0.01 µL to 0.1 mL (volume of viral stock
added per mL = “replication factor”). Use at least two dilutions that differ by
10-fold and will give a countable number of colonies after selection. Add
polybrene to a final concentration of 8 µg/mL and incubate cells at 37°C for
1–3 h.
c. Add medium to dilute the polybrene to 2 µg/mL and incubate at least 2–3 d.
d. Split the infected cells at 1:10 and 1:20 (1/10 or 1/20 = “fraction of infected
cells plated”) into selection medium containing 400 µg/mL of G418 and
incubate for 3 d. Change to fresh medium containing G418 and incubate
until colonies are obvious (7–10 d). Count colonies before they spread and
calculate titer as:
G418–resistant CFU/mL = no. of colonies

virus vol. × replication factor × fraction of infected cells plated
in which CFU = colony-forming units.
28 Goldring
3.2. Retroviral Infection of Chondrocytes

1. Plate freshly isolated chondrocytes at a density of 2.5 × 104 per cm2 in flasks or
dishes. Culture at 37°C for 3–4 d, with medium changes every 2 d, or until the
cells begin to divide rapidly and have covered more than 50% of the plate (see
Note 5).
2. At 24 and 3 h prior to infection, remove the spent medium from chondrocyte
cultures and add fresh growth medium to stimulate cell division. Hyaluronidase
may be added during these medium changes to facilitate the infection efficiency
(see Note 6).
3. Begin infection by removing the medium and adding 5 mL of freshly thawed
virus-containing medium supplemented with polybrene at 8 µg/mL to promote
the binding of virus to the cells. Replace the dishes in the 37°C incubator and
incubate for 4–6 h or overnight (see Note 7).
4. Remove the polybrene-containing medium and replace with fresh growth medium.
Incubate for at least 24 h, until the cultures become confluent, or for at least one
cell division cycle.
3.3. Selection and Establishment of Immortalized Chondrocytes

1. Passage confluent cultures with a split ratio of 1:2, and begin selection with G418
the following day. Cultures infected with tsSV40-TAg may be transferred to a
32°C incubator at this point.
2. Remove the medium and add fresh growth medium containing G418 at 300 µg/
mL (see Note 3). Continue selection for 3–4 wk, feeding the cultures three times
per week with fresh growth medium supplemented with G418. During selection
a percentage of the cells will die, substantially reducing the culture density. After
4–8 wk, however, G418-resistant cells should repopulate the dishes, and when
the cultures become confluent, they should be passaged and maintained in fresh
growth medium containing G418 at 200 µg/mL.
3. Establish cultures by passaging more than 50 times in growth medium without
G418, ensuring that the cultures are allowed only to reach “confluence” and not
overgrow, with the appearance of floating cells (see Note 8). Cultures are main-
tained as nonclonal populations or cloned by limiting dilution after “crisis” (see
Note 9). After stable growth for approx 25 or more population doublings (same
as passage number in cultures split from one to two dishes), a gradual decline in
growth rate is observed with a progressive failure of the cells to grow to
confluence. This crisis period is marked by morphological changes in a portion
of the cells that are destined to die and the accumulation of cellular debris. Some
cultures undergo a latency period of a few weeks, after which the rate of cell
division returns to that prior to crisis.
4. Verify the presence of the immortalizing antigen in established, G418-resistant
chondrocyte cell lines by immunocytochemistry and/or by Western blotting
using specific antibodies against the immortalizing antigen, i.e., SV40-TAg,
HPV16-E6, HPV16-E7, or hTERT. The mRNAs encoding these antigens may
also be analyzed by reverse transcriptase polymerase chain reaction (RT-PCR).
Fig. 1. Morphologies of immortalized human chondrocytes in monolayer culture.

(A) T/C-28a4, (B) C-28/I2, (C) C-20/A4, and (D) tsT/AC62 cells were cultured to
confluency and photographed using phase-contrast microscopy. The T/C-28a4, C-28/
I2, and C-20/A4 cell lines were derived from primary cultures of juvenile human cos-
tal chondrocytes by immortalization with a retroviral or plasmid vector encoding
SV40-TAg (15). The tsT/AC62 cell line was derived from primary cultures of adult
human articular chondrocytes by immortalization with a temperature-sensitive mutant
of SV40-TAg (16). Note that the T/C-28 cells are small and rounded. The tsT/AC62
cells become more fibroblastic during passage and may be “redifferentiated” about
every 12 passages by culture in alginate or in suspension for 1–2 wk. (Reproduced
with permission from ref. 30.)
Telomerase activity in hTERT-immortalized cells is analyzed by the telomerase

repeat amplification protocol (TRAP). Telomeric length, which decreases with
senescence and increases in hTERT-immortalized cells, is detected by Southern
blotting using genomic DNA digested with the restriction enzymes RsaI and MspI
or HinfI and the 32P end-labeled telomeric probe (TTAGGG)3 (25). Mean
telomeric length is determined by densitometric analysis of autoradiographs of
the Southern blots (26).
3.4. Characterization of Established Chondrocyte Cell Lines

1. Morphology: if cultures are passaged at confluence at split ratios that permit
appropriate densities (see Note 8), chondrocyte cell lines will maintain the typi-
cal rounded to polygonal morphology in monolayer culture, although the size
and volume may vary, as shown in Fig. 1.
30 Goldring
Fig. 2. Representative growth curves from established immortalized chondrocyte

cell lines expressing different forms of SV40-TAg. Cells were seeded into multiwell
plates at a density of 2.5 × 104 cells/cm2 and cultured in DMEM/F-12 supplemented
with 10% FCS for the indicated time periods with medium changes every 3 d. The T/C-
28a4, C-28/I2, and C-20/A4 cell lines are grown at 37°C. The tsT/AC62 cell line is
grown at 32°C, the permissive temperature for expression of tsTAg. (Reproduced with
permission from ref. 30.)
2. Growth kinetics: trypsinize the cells from a 10-cm dish and wash with PBS. Plate
cells in 12-well plates at a concentration of 50,000 cells/well in 2 mL of growth
medium. At intervals of 48 h, trypsinize duplicate wells from each plate and deter-
mine the cell counts using a hemocytometer or Coulter counter. Cell counts should
be obtained until the cultures reach confluence. The proliferative capacity may
also vary from one cell line to another, depending on the source of cartilage and
the technique of immortalization (Fig. 2).
3. Phenotypic characterization (see Note 10): it is necessary to verify the presence
of the definitive markers of the phenotype in chondrocyte cell lines after selec-
tion and to screen frequently once they are established. Phenotype can be moni-
tored conveniently by the analysis of mRNAs encoding COL2A1 and aggrecan
using sensitive RT-PCR methods (see Note 11), as described previously (27–30)
and in Chapters 6 and 7, in this volume. As shown in Fig. 3, the expression of
COL2A1 mRNA is lower in proliferating chondrocyte cell lines, where it is sup-
pressed by 10% FCS, and can be enhanced in the presence of an insulin-contain-
ing serum substitute, such as Nutridoma-SP or ITS+. The expression of aggrecan
mRNA is not coordinately regulated with COL2A1 mRNA and is less susceptible
to changes in the culture conditions (see Note 12). Experimental incubations are
performed as follows:
a. Passage cells into 6-well culture plates at 2.5–5 × 105 cells/well in DMEM/F-
12 containing 10% FCS and allow the cells to grow to 50–95% confluence.
Fig. 3. Matrix gene expression in immortalized chondrocyte cell lines. Total RNA
was isolated and analyzed by RT-PCR for expression of COL2A1, aggrecan, and
GAPDH mRNAs. Lane 1, DNA ladder; lane 2, primary adult articular chondrocytes;
lane 3, T/C-28a4 cultured in medium with 10% FCS; lane 4, T/C28a4 in serum-free
medium with 1% Nutridoma-SP; lane 5, C-28/I2 in 1% Nutridoma-SP; lane 6, C-28/I2
in 10% FCS; lane 7, C-20/A4 in 10% FCS; lane 8, tsT/AC62 in 10% FCS at 32°C.
Note that some cell lines are susceptible to a loss of the COL2A1 phenotype in serum-
containing growth medium. (Reproduced with permission from ref. 30.)
b. Remove culture medium, wash with PBS, and replace with DMEM/F-12 con-
taining 1% Nutridoma-SP or 1% ITS+.
c. Incubate overnight for up to 24 h, add treatments in a small volume without
medium change, and continue incubations for the times desired.
4. Notes
1. Batches of serum should be tested and selected on the basis of the capacity to
support expression of chondrocyte-specific matrix gene expression. High capac-
ity to induce cell proliferation is not necessarily associated with the ability to
maintain phenotype.
2. Human chondrocytes, similar to other human cell types, are not as easily immor-
talized with retroviruses as their mouse counterparts. Amphotropic packaging
cell lines that have a broad host range are used for the production of helper virus-
free stocks that will infect human cells (24).
3. The appropriate G418 concentration should be determined empirically for each
cell line prior to transfection. For most packaging cell lines, 300–400 µg/mL is
optimal for death of nontransfected cells. The neomycin-sensitive cells should
stop growing within 24 h, and cell death should be evident within 1 wk. Since
multiplying cells are susceptible to G418, frequent changes of medium contain-
ing fresh G418 should be applied. Once cell death is evident, the G418 concen-
tration can be changed to 200 µg/mL.
32 Goldring
4. Safety considerations in handling viral stocks: Special care should be taken to

prevent skin contamination and aerosol of virus particles. Spent medium, pipets,
vacuum flasks, and any other materials in contact with virus-containing media
should be treated with a 10% solution of bleach or handled as recommended by
institutional standards.
5. After initial plating of the primary cultures, the chondrocytes require 2–3 d before
they have settled down and spread out completely. Although the cultures may
continue to express chondrocyte phenotype (e.g., type II collagen and aggrecan
mRNAs) for several weeks, expression of nonspecific collagens I and III may
begin as early as d 7 after isolation. Thus, it is recommended that infection with
the immortalizing retrovirus be carried out between d 4 and 7 after isolation. Since
cells must be actively dividing in order to be infected efficiently with retroviruses,
the culture medium should be changed 24 and 3 h prior to addition of the retroviral
supernatants.
6. Hyaluronidase added before and/or during transfections has been shown to increase
transfection efficiencies in chondrocytes (31–33) and also to facilitate retroviral
gene transfer in other cell types by degrading hyaluronan and GAGs in the pericel-
lular coat and cell surface (34).
7. Since viral activity is depleted after 4–6 h of incubation at 37°C, it is not neces-
sary to leave the viral supernatants on the chondrocytes overnight except as a
convenience.
8. The expansion of immortalized chondrocytes in monolayer culture in medium
containing 10% FCS favors a loss of phenotype. Established cultures should not
be plated at densities lower than those resulting from a split ratio of 1:2 for adult
articular chondrocytes and 1:5–1:10 for more rapidly growing chondrocytes, such
as juvenile costal chondrocytes. They should be passaged immediately upon reach-
ing confluence, but not more than twice weekly. Since prolonged passaging of
immortalized cell lines will select for rapidly growing cells, the cell lines should
be expanded as soon as they are established and frozen down in liquid nitrogen.
9. Cultures are maintained as nonclonal populations, at least until they are esta-
blished, in order to preserve potential cellular interactions among clonal cell types
that might aid in survival and the preservation of phenotype. If the cells are cloned
immediately after G418 selection, when immortalized cells have not completely
repopulated the dish, survival is poor compared with other cell types. Greater
success can be achieved if cloning by limiting dilution is performed on esta-
blished cultures after crisis.
10. A note of caution is that immortalized human chondrocyte cell lines should not
serve as a replacement for primary cultures. Because of the stable proliferative
capacities of established cell lines, the cultures may be expanded to produce suf-
ficient numbers of cells for selected types of experiments (see Chapter 4).
11. COL2A1 mRNA, detected by RT-PCR, is the most sensitive marker of the chon-
drocyte phenotype, since it is lost more easily than aggrecan mRNA.
12. Further notes on culture of human chondrocytes and on chondrocyte immortaliza-
tion may be found in previous volumes of Methods in Molecular Medicine (27,28).
Acknowledgments
This work was supported in part by NIH grants AR45378 and AG22021 and
a Biomedical Science Grant from the Arthritis Foundation.
References
1.
1 Goldring, M. B. (2000) The role of the chondrocyte in osteoarthritis. Arthritis
Rheum. 4, 1916–1926.
2.
2 Goldring, M. B., Sandell, L. J., Stephenson, M. L., and Krane, S. M. (1986) Immune
interferon suppresses levels of procollagen mRNA and type II collagen synthesis in
cultured human articular and costal chondrocytes. J. Biol. Chem. 261, 9049–9056.
3.
3 Brittberg, M., Lindahl, A., Nilsson, A., Ohlsson, C., Isaksson, O., and Peterson,
L. (1994) Treatment of deep cartilage defects in the knee with autologous chon-
drocyte transplantation. N. Engl. J. Med. 331, 889–895.
4.
4 Alema, S., Tato, F., and Boettiger, D. (1985) Myc and src oncogenes have comple-
mentary effects on cell proliferation and expression of specific extracellular matrix
components in definitive chondroblasts. Mol. Cell. Biol. 5, 538–544.
5.
5 Gionti, E., Pontarelli, G., and Cancedda, R. (1985) Avian myelocytomatosis virus
immortalizes differentiated quail chondrocytes. Proc. Natl. Acad. Sci. USA 82,
2756–2760.
6.
6 Horton, W. E., Jr, Cleveland, J., Rapp, U., et al. (1988) An established rat cell line
expressing chondrocyte properties. Exp. Cell Res. 178, 457–468.
7.
7 Thenet, S., Benya, P. D., Demignot, S., Feunteun, J., and Adolphe, M. (1992)
SV40-immortalization of rabbit articular chondrocytes: alteration of differenti-
ated functions. J. Cell. Physiol. 150, 158–167.
8.
8 Mallein-Gerin, F. and Olsen, B. R. (1993) Expression of simian virus 40 large T
(tumor) oncogene in chondrocytes induces cell proliferation without loss of the
differentiated phenotype. Proc. Natl. Acad. Sci. USA 90, 3289–3293.
9.
9 Grigoriadis, A. E., Heersche, J. N. M., and Aubin, J. E. (1988) Differentiation of
muscle, fat, cartilage, and bone from progenitor cells present in a bone-derived
clonal cell population: effect of dexamethasone. J. Cell Biol. 106, 2139–2151.
10.
10 Bernier, S. M. and Goltzman, D. (1993) Regulation of expression of the chondro-
genic phenotype in a skeletal cell line (CFK2) in vitro. J. Bone Miner. Res. 8,
475–484.
11.
11 Lefebvre, V., Garofalo, S., and deCrombrugghe, B. (1995) Type X collagen gene
expression in mouse chondrocytes immortalized by a temperature-sensitive sim-
ian virus 40 large tumor antigen. J. Cell Biol. 128, 239–245.
12.
12 Mataga, N., Tamura, M., Yanai, N., et al. (1996) Establishment of a novel chon-
drocyte-like cell line derived from transgenic mice harboring the temperature-
sensitive simian virus 40 large T-antigen. J. Bone Miner. Res. 11, 1646–1654.
13. Benoit, B., Thenet-Gauci, S., Hoffschir, F., Penformis, P., Demignot, S., and
Adolphe, M. (1995) SV40 large T antigen immortalization of human articular
chondrocytes. In Vitro Cell. Dev. Biol. 31, 174–177.
14.
14 Steimberg, N., Viengchareun, S., Biehlmann, F., et al. (1999) SV40 large T anti-
gen expression driven by col2a1 regulatory sequences immortalizes articular
34 Goldring
chondrocytes but does not allow stabilization of type II collagen expression. Exp.
Cell Res. 249, 248–259.
15.
15 Goldring, M. B., Birkhead, J. R., Suen, L.-F., et al. (1994) Interleukin-1β-modu-
lated gene expression in immortalized human chondrocytes. J. Clin. Invest. 94,
2307–2316.
16.
lar chondrocytes maintain cartilage-specific phenotype and responses to interleukin-
1β. Arthritis Rheum. 43, 2189–2201.
17.
17 Grigolo, B., Roseti, L., Neri, S., et al. (2002) Human articular chondrocytes immor-
talized by HPV-16 E6 and E7 genes: maintenance of differentiated phenotype under
defined culture conditions. Osteoarthritis Cartilage 10, 879–889.
18.
18 Piera-Velazquez, S., Jimenez, S. A., and Stokes, D. (2002) Increased life span of
human osteoarthritic chondrocytes by exogenous expression of telomerase. Arthri-
tis Rheum. 46, 683–693.
19.
19 Miller, A. D. and Buttimore, C. (1986) Redesign of retrovirus packaging cell
lines to avoid recombination leading to helper virus production. Mol. Cell Biol.
6, 2895–2902.
20.
20 Ory, D. S., Neugeboren, B. A., and Mulligan, R. C. (1996) A stable human-derived
packaging cell line for production of high titer retrovirus/vesicular stomatitis virus
G pseudotypes. Proc. Natl. Acad. Sci. USA 93, 11,400–11,406.
21.
21 Jat, P. S., Cepko, C. L., Mulligan, R. C., and Sharp, P. A. (1986) Recombinant
retroviruses encoding simian virus 40 large T antigen and polyomavirus large and
middle T antigens. Mol. Cell Biol. 6, 1204–1217.
22.
22 Jat, P. S. and Sharp, P. A. (1989) Cell lines established by a temperature-sensitive
simian virus 40 large-T-antigen gene are growth restricted at the nonpermissive
temperature. Mol. Cell. Biol. 9, 1672–1681.
23.
23 Harley, C. B. (2002) Telomerase is not an oncogene. Oncogene 21, 494–502.
24. Cepko, C. L. and Pear, W. (2002) Introduction of DNA into mammalian cells, in
Short Protocols in Molecular Biology, 5th Ed., Vol. 1, (Ausebel, F. M., Brent, R.,
Kingston, R. E., et al., eds.), John Wiley, New York, NY, pp. 9-1–9-77.
25.
25 Bodnar, A. G., Ouellette, M., Frolkis, M., et al. (1998) Extension of life-span by
introduction of telomerase into normal human cells. Science 279, 349–352.
26.
26 Harley, C. B., Futcher, A. B., and Greider, C. W. (1990) Telomeres shorten dur-
ing ageing of human fibroblasts. Nature 345, 458–460.
27.
27 Goldring, M. B. (1996) Human chondrocyte cultures as models of cartilage-spe-
cific gene regulation, in Methods in Molecular Biology: Human Cell Culture Pro-
tocols (Jones, G. E., ed.), Humana, Totowa, NJ, pp. 217–231.
28. Robbins, J. R. and Goldring, M. B. (1998) Methods for preparation of immortal-
ized human chondrocyte cell lines, in Methods in Molecular Medicine: Tissue
Engineering Methods and Protocols (Morgan, J. R. and Yarmush, M. L., eds.),
29
29. Kokenyesi, R., Tan, L., Robbins, J. R., and Goldring, M. B. (2000) Proteoglycan
production by immortalized human chondrocyte cell lines cultured under condi-
tions that promote expression of the differentiated phenotype. Arch. Biochem.
Biophys. 383, 79–90.
30.
30 Loeser, R. F., Sadiev, S., Tan, L., and Goldring, M. B. (2000) Integrin expression
by primary and immortalized human chondrocytes: evidence of a differential role
for α1β1 and α2β1 integrins in mediating chondrocyte adhesion to types II and
VI collagen. Osteoarthritis Cartilage 8, 96–105.
31.
31 Lu Valle, P., Iwamoto, M., Fanning, P., Pacifici, M., and Olsen, B. R. (1993)
Multiple negative elements in a gene that codes for an extracellular matrix pro-
tein, collagen X, restrict expression to hypertrophic chondrocytes. J. Cell Biol.
121, 1173–1179.
32.
32 Viengchareun, S., Thenet-Gauci, S., Steimberg, N., Blancher, C., Crisanti, P., and
Adolphe, M. (1997) The transfection of rabbit articular chondrocytes is indepen-
dent of their differentiation state. In Vitro Cell Dev. Biol. Anim. 33, 15–17.
33.
33 Madry, H. and Trippel, S. B. (2000) Efficient lipid-mediated gene transfer to articu-
lar chondrocytes. Gene Ther. 7, 286–291.
34. Batra, R. K., Olsen, J. C., Hoganson, D. K., Caterson, B., and Boucher, R. C. (1997)
Retroviral gene transfer is inhibited by chondroitin sulfate proteoglycans/gly-
cosaminoglycans in malignant pleural effusions. J. Biol. Chem. 272, 11,736–11,743.
Chondrocyte Cell Lines As Culture Models 37
4
Culture of Immortalized Chondrocytes
and Their Use As Models of Chondrocyte Function
Mary B. Goldring
Summary
Immortalization of chondrocytes increases life span and proliferative capacity but does not
necessarily stabilize the differentiated phenotype. Expansion of chondrocyte cell lines in con-
tinuous monolayer culture may result in the loss of phenotype, particularly if high cell density
is not maintained. This chapter describes strategies for maintaining or restoring differentiated
phenotype in established chondrocyte cell lines involving culture in serum-free defined cul-
ture medium, in suspension over agarose or polyHEMA, or within alginate or collagen scaf-
folds. Chondrocyte cell lines have been used successfully to develop reproducible models for
studying the regulation of gene expression in experiments requiring large numbers of cells.
Thus, approaches for studying transcriptional regulation by transfection of promoter-driven
reporter genes and cotransfection of expression vectors for wild-type or mutant proteins are
also described.
Key Words: Immortalized chondrocytes; type II collagen; aggrecan; monolayer culture;

suspension culture; alginate; agarose; polyHEMA; three-dimensional (3D) scaffolds; collagen
scaffolds; ascorbate; Nutridoma; transfections; luciferase reporter plasmids; green fluorescent
protein (GFP) reporter plasmids; adenoviral-mediated expression; luciferase assay.
1. Introduction
Simian virus 40 large T antigen (SV40-TAg), human papillomavirus type
16 early function genes E6 and E7 (HPV-16 E6/E7), and telomerase have been
used with varying degrees of success for immortalizing human chondrocytes,
as described in Chapter 3. However, a general observation is that phenotypic
stability is lost during serial subculture of immortalized chondrocytes in
monolayer culture. Clonal expansion usually results in type II collagen-negative
cell lines that express type I and type III collagens in monolayer culture (1,2).
The loss of phenotype may be reversible if established monolayer cultures are
37
38 Goldring
maintained as nonclonal populations and selected by passaging through

suspension cultures (3,4). Furthermore, phenotype can be restored by transfer
to three-dimensional (3D) culture in alginate (4) or hyaluronan (5) or to sus-
pension culture on poly-2-hydroxyethyl-methacrylate (polyHEMA)-coated
dishes (6). Interestingly, long-term pellet culture of a clonal cell line of tsTAg-
immortalized human articular chondrocytes at 37°C, the nonpermissive tem-
perature for proliferation, induced expression of type X collagen, a
hypertrophic chondrocyte marker not found in normal articular cartilage (7).
Success in establishing immortalized human chondrocyte cell lines that can
be used as models for studying cartilage breakdown and repair depends not
only on the source and developmental state of the tissue from which the cell
lines are established, but also more importantly on the ability of the culture
system to support the differentiated phenotype. Regardless of the immortaliza-
tion strategy, however, matrix protein synthesis and secretion in general and
chondrocyte-specific phenotype in particular may be reduced after selection
and expansion in monolayer culture. Thus, short-term incubation in serum-free
defined medium supplemented with an insulin-containing serum substitute has
been used as an experimental strategy (3–5). This chapter focuses on strategies
for culture and characterization that are applicable to any chondrocyte cell line
and also describes approaches for using immortalized chondrocytes as repro-
ducible models for evaluating chondrocyte functions.
2. Materials
2.1. Culture Reagents
1. Growth medium for chondrocytes: Dulbecco’s modified Eagle’s medium
(DMEM)/Ham’s F-12, 1:1 (v/v). Mix equal volumes of DMEM, high glucose,
with L-glutamine and Ham’s F-12, with L-glutamine (Cambrex, Walkersville, MD),
both without HEPES, to give final concentrations of 3.151 g/L glucose, L-glutamine
(365 mg/L)/L-glutamic acid (7.36 mg/L), and sodium pyruvate (110 mg/L). Add
10% FCS immediately before use.
2. Dulbecco’s phosphate-buffered Ca2+- and Mg2+-free saline (PBS).
3. Trypsin-EDTA solution: 0.05% trypsin and 0.02% ethylenediamine tetraacetic
acid (EDTA) in Hanks’ balanced salt solution without Ca2+ and Mg2+ (Life Tech-
nologies, Gaithersburg, MD).
4. Hanks’ balanced salt solution with Ca2+ and Mg2+ (HBSS; Life Technologies).
5. Serum substitutes for experimental incubations of immortalized chondrocytes:
Nutridoma-SP (Roche Applied Science) is provided as sterile concentrate (100X,
pH 7.4; storage at 15–25°C protected from light). Dilute 1:100 (v/v) with sterile
DMEM/Ham’s F-12, without FCS, and use immediately. ITS+ (BD Biosciences)
is an alternative serum substitute.
Except where specified, cell culture reagents can be obtained from a num-
ber of different suppliers, including Life Technologies, Cambrex, Sigma (St.
Louis, MO), or Intergen (Purchase, NY). Reserve serum testing, which is

offered by these suppliers, is recommended for selection of lots of FCS that
maintain chondrocyte phenotype (see Note 1). FCS and trypsin-EDTA are
stored at –20°C but should not be refrozen after thawing for use.
2.2. Culture Systems for Immortalized Chondrocytes
1. Agarose-coated dishes (8).
a. Weigh out 10 mL of high melting point agarose in an autoclavable bottle and
add 100 mL of dH2O.
b. Autoclave with cap tightened loosely, allow to cool to approx 55°C, and pipet
quickly into culture dishes (1 mL/3.5-cm well of 6-well plate, 3 mL/6-cm
dish, or 9 mL/10-cm dish).
c. Allow the gel to set at 4°C for 30 min and wash the surface two to three times
with PBS.
d. Plates may be used immediately or wrapped tightly with plastic or foil to
prevent evaporation and stored at 4°C.
2. Dishes coated with polyHEMA (9).
a. Prepare a 10% (w/v) solution by dissolving 5 g of polyHEMA (BD Bio-
sciences) in 50 mL of ethanol in a sterile capped bottle or centrifuge tube.
b. Leave overnight at 37°C with gentle shaking to dissolve polymer completely.
c. Centrifuge the viscous solution for 30 min at 2000g to remove undissolved
particles.
d. Layer the polyHEMA solution on dishes at 0.3 mL/well of 6-well plate or
0.9 mL/6-cm dish and leave with lids in place to dry overnight in a tissue
culture hood.
e. Expose open dishes to bactericidal ultraviolet light for 30 min to sterilize.
3. Alginate (Keltone LVCR, NF; ISP Alginates, San Diego, CA). Low viscosity (LV)
alginate is generally used. Request LVCR for more highly purified preparation.
a. Prepare 1.2% (w/v) solution of alginate in 0.15 M NaCl.
b. Dissolve alginate in a 0.15 M NaCl solution, heating the solution in a micro-
wave oven until it just begins to boil. Swirl and heat again two or three times
until the alginate is dissolved completely. (Caution: Do not autoclave.) Allow
the solution to cool to about 37°C and sterile filter. Filtering when warm per-
mits the viscous solution to pass through the filter.
c. Prepare 102 mM CaCl2 and 0.15 M NaCl solutions in tissue culture bottles
and autoclave.
d. Prepare 55 mM Na citrate, 0.15 M NaCl, pH 6.0, sterile filter, and store at
4°C. Make fresh weekly.
4. 3D scaffolds: several types of scaffolds are available commercially, including
the following:
a. Gelfoam ® (Pharmacia & Upjohn, Kalamazoo, MI), sterile absorbable
collagen sponge, purchased as size 12–7 (2 cm × 6 cm × 7 mm; cat. no. 09-
0315-03, box of 12).
40 Goldring
b. BD™ Three Dimensional Collagen Composite Scaffold (BD Biosciences):

contains a mixture of bovine type 1 and type III collagens and is provided as
3D scaffolds with 48-well plates.
c. BD™ Three Dimensional OPLA® Scaffold (BD Biosciences): contains a syn-
thetic polymer synthesized from D,D,-L,L polylactic acid and is provided as 3D
scaffolds with 48-well plates.
5. Recovery of cells from scaffolds:
a. Cell lysis solution: 0.2% v/v Triton X-100, 10 mM Tris-HCl, pH 7.0, 1 mM
EDTA.
b. Collagenase solution: 0.03% (w/v) collagenase (Worthington, Freehold, NJ)
in HBSS.
c. RNA extraction kits: TRIzol® reagent (Life Technologies) or RNeasy Mini
Kit (Qiagen).
2.3. Immortalized Chondrocyte Cell Lines

As Models for Studying Chondrocyte Function
1. EndoFree Plasmid Maxi Kit (Qiagen).
2. Lipid-based transfection reagent such as LipofectAMINE PLUS™ Reagent (Life
Technologies) or FuGENE 6 (Roche Applied Science, Indianapolis, IN).
3. Serum-free culture medium for transfections: Opti-MEM (Life Technologies) or
DMEM/F-12 (test for optimal transfection efficiency).
4. Passive Lysis Buffer (Promega).
5. Coomassie Plus Protein Assay Reagent (Pierce, Rockford, IL).
6. Luciferase Assay System (Promega, Madison, WI).
7. Dual-Luciferase Reporter Assay (Promega) with the pRL-TK Renilla luciferase
control vector.
8. Adenovirus producer cell line: 293 (ATTC CRL 1573; transformed primary hu-
man embryonic kidney).
3. Methods
Immortalization strategies and approaches for selection and expansion of
chondrocyte cell lines are described in Chapter 3. The methods described be-
low outline: (1) culture systems for immortalized chondrocytes, and (2) the use
of established cell lines as models for studying chondrocyte function.
3.1. Culture Systems for Immortalized Chondrocytes

Immortalization of a number of human somatic cell types, including dermal
fibroblasts, osteoblasts, endothelial cells, and epithelial cells, using SV40-TAg,
HPV-16 E6/E7, or hTERT has been shown to increase life span and prolifera-
tive capacity, but it does not necessarily preserve the differentiated phenotype.
Since proliferating chondrocytes in monolayer culture are particularly suscep-
tible to loss of phenotype during prolonged culture, it is necessary to use cul-
ture conditions that maintain differentiated chondrocyte features or that permit

redifferentiation (see Note 2).
3.1.1. Monolayer Cultures
1. Immortalized chondrocytes are cultured routinely in 10-cm dishes and passaged
immediately upon reaching confluence. Adult articular chondrocytes are pas-
saged every 7–10 d and more rapidly growing cultures, such as juvenile costal
chondrocytes, are passaged every 4–6 d.
2. For passaging, remove culture medium by aspiration with a sterile Pasteur pipet
attached to a vacuum flask and wash with PBS (Ca2+- and Mg2+-free). Add trypsin-
EDTA solution (not trypsin alone) at 1.5 mL/10-cm dish to cover cell layer, and
immediately remove 1 mL. Leave at room temperatures with periodic gentle shak-
ing of dish. Check in microscope to ensure that all cells have come off the plate in
a single cell suspension (see Note 3). If significant numbers of cells remain at-
tached, continue the incubation for a longer time (20 min or less) at 37°C and/or
gently scrape the cell layer with a sterile plastic scraper, Teflon policeman, or
syringe plunger.
3. Bring up cells in 5- or 10-mL pipet containing culture medium, aspirating in and
out several times to break up cell clumps. Transfer to a sterile conical 15- or 50-mL
polypropylene tube and perform cell counts to determine the split ratio required
(usually 1 dish into 2 [1:2] for adult articular chondrocytes, or 1 dish into 5, 8, or
10 for more rapidly growing juvenile chondrocytes). Distribute equal volumes of
the cell suspension in dishes or wells that already contain culture medium, rocking
plates back and forth (not swirling) immediately after each addition to ensure uni-
form plating density throughout the plate.
4. For experiments, plate cells in 12-well plates at a concentration of 100,000 cells/
well in 2 mL of growth medium or in 6-well plates at 250,000 cells/well in 3 mL
of medium. Remove growth medium 1–3 d after passaging, when the cultures are
subconfluent. Add serum-free medium containing a serum substitute (see Note 4),
such as Nutridoma-SP, followed immediately or 1–24 h later by the test agent of
interest. Continue incubation at 37°C for short-term time courses of 0.25–24 h or
longer time-courses of 24–72 h.
3.1.2. Fluid Suspension Cultures

on Agarose- or polyHEMA-Coated Dishes
1. Trypsinize monolayer cultures, spin down cells, wash with PBS, centrifuge, and
resuspend in culture medium containing 10% FCS at 1 × 106 cells/mL.
2. Transfer chondrocyte suspension to dishes that have been coated with 1% agar-
ose or with 0.9% polyHEMA and culture for 2–4 wk. The cells first form large
clumps that begin to break up after 7–10 d and eventually form single-cell sus-
pensions.
3. Change the medium weekly by carefully removing the medium above settled cells
while tilting the dish, centrifuging the remaining suspended cells and replacing
them in the dish after resuspension in fresh culture medium.
42 Goldring
4. To recover cells for direct experimental analysis, for redistribution in agarose- or

polyHEMA-coated wells, or for culture in monolayer, transfer the cell suspen-
sion to 15- or 50-mL conical tubes, gently washing the agarose surface at least
twice with culture medium to recover remaining cells; then spin down and resus-
pend cells in an appropriate volume of culture medium for plating or in extrac-
tion buffer for subsequent experimental analysis.
3.1.3. Alginate Bead Cultures (see Note 5)

1. Trypsinize several 10-cm plates and wash the cells with PBS. Determine the cell
count with a hemocytometer and pellet the cells. Resuspend the pellet in a 1.2%
solution of alginate, 0.15 M NaCl at a concentration of 4 × 106 cells/mL.
2. Express the alginate suspension in a dropwise manner through a 10-mL syringe
equipped with a 25-gage needle into a 50-mL centrifuge tube containing 40 mL
of 102 mM CaCl2. Allow the beads to polymerize in the CaCl2 solution for 10
min and wash twice with 25 mL of 0.15 M NaCl. The alginate beads should not
be washed in PBS, as they will become cloudy.
3. Resuspend the beads (7–15 beads/mL) in 20 mL of growth medium supplemented
with 25 µg/mL Na ascorbate (see Note 6), and decant the beads to a 10-cm dish.
As many as 150 beads (approx 9 × 106 cells) can be cultured in a single dish with
medium changes every 3 d.
4. To recover cells from alginate, carefully aspirate the medium from the cultures
and wash twice with PBS. Depolymerize the alginate by adding three volumes of
a solution of 55 mM Na citrate/0.15 M NaCl and incubate at 37°C for 10 min.
Aspirate the solution over the surface of the dish several times to dislodge adher-
ent cells (the cells are sticky) and transfer the suspension to a 50-mL centrifuge
tube.
5. Because the solution is quite viscous, centrifuge the cells at 2000g for a mini-
mum of 10 min to pellet the cells completely. Wash the cells twice with PBS
before using them for further analysis.
3.1.4. Culture on 3D Scaffolds (see Note 7)
1. Gelfoam: use a sterile scalpel blade to cut into pieces of 1 × 1 × 0.5 cm3 and place
in wells of sterile 6-well plates. Inoculate by dropping 50 µL of growth medium
containing 106 cells on each sponge. Place in incubator for 1.5–2 h, and then add
100 µL medium and culture for an additional 1–3 h. Add medium to cover and
continue incubation overnight or longer.
2. BD 3D Collagen Composite or OPLA Scaffolds: place scaffolds (0.5 cm3) in the
48-well plates provided, in 96-well plates, or in other plates as required. Seed
scaffolds by dropping 100 µL of growth medium containing 1–5 × 104 cells. Incu-
bate for 1 h, add 150 µL of medium to each scaffold, and incubate for 1.5–3 h.
Add medium as required for further culture and experimental conditions.
3. Recovery of cells from scaffolds for analysis:
a. Prepare cell lysates for DNA analysis using 250–500 µL of cell lysis solution
(0.2% v/v Triton X-100, 10 mM Tris-HCl, pH 7.0, 1 mM EDTA) per scaffold
in 1.5-mL tube. Freeze samples at –70°C and subject to two freeze/thaw

cycles, thawing at room temperature for 45–60 min. Break up scaffolds with
pipet tip, centrifuge, and transfer lysates to fresh tubes. The cell lysates may
be analyzed using the Picogreen Assay Kit according to the manufacture’s
protocol (Molecular Probes).
b. Recover cells for RNA extraction and other analyses by treatment of minced
scaffolds with 0.03% (w/v) collagenase in HBSS for 10–15 min at 37°C. Col-
lect cells by centrifugation, wash with PBS, and extract total RNA using
TRIzol reagent or RNeasy Mini Kit.
3.2. Immortalized Chondrocyte Cell Lines

As Models for Studying Chondrocyte Function
Immortalized chondrocyte cell lines serve as reproducible models to study
the expression of chondrocyte-specific matrix genes and to examine the ef-
fects of cytokines and growth/differentiation factors on chondrocyte pheno-
typic markers, as shown in Figs. 1 and 2. In 3D cultures, it is also useful to
characterize the proteoglycans synthesized (4,10,11) (see also Chapters 14
and 15), stain the matrix components with alcian blue (4,10), toluidine blue,
or other dyes, or perform immunohistochemistry using specific antibodies
against these proteins (3) (see Chapters 17 and 18). Other features of the
chondrocyte phenotype include the expression of the three members of the
SRY-type HMG box (SOX) family of transcription factors, L-Sox5, Sox6,
and Sox9 (12), which are required for chondrocyte differentiation during
development. The capacity to respond to interleukin-1β (IL-1β) by increas-
ing expression of cyclooxygenase-2 (COX-2) and matrix metalloproteinase-
13 (MMP-13) is somewhat dependent on the differentiated phenotype (13).
Integrin profiles should be consistent with those expressed on normal
chondrocytes but may also reflect the types of integrins found on proliferat-
ing cells (14). However, immortalized chondrocytes should not be consid-
ered as substitutes for primary chondrocytes, which should be used to validate
key findings.
Chondrocyte cell lines have served most beneficially as models to delineate
the cytokine- and growth factor-induced signaling pathways and transcription
factors involved in the regulation of gene expression. They may be used to
study the effects of selective protein kinase inhibitors, as described (4,13),
employing techniques described in Chapter 20. Because they can be cultured
in sufficient numbers, they are convenient models for examining transiently
or stably expressed reporter genes or recombinant wild-type and mutant
proteins (13,15–20) and for examining specific DNA binding activities in
nuclear extracts (12,21). They may also be used to test approaches for gene
therapy using viral vectors (22) and RNA interference (RNAi) strategies (23).
44 Goldring
Fig. 1. RT-PCR analysis of mRNAs expressed by immortalized human

chondrocytes. The tsT/AC62 cells were passaged into 6-well plates in DMEM/F-12
containing 10% FCS and incubated at 32°C for 4 d, at which time the medium was
changed. On d 5, IL-1β (100 pg/mL) or hBMP-2 (100 ng/mL) was added alone or in
combination, and incubations were continued until d 7. Total RNA was extracted using
TRIzol reagent and mRNAs encoding COL2A1, aggrecan, matrix metalloproteinase
(MMP)-1, MMP-3, phospholipase A2 (PLA2), and GAPDH were analyzed by RT-
PCR as described in Robbins et al. (4). The DNA ladder is shown on the left. Lane 1,
untreated control; lane 2, IL-1β; lane 3, BMP-2; and lane 4, IL-1β and BMP-2.
Plasmid vectors, in which the expression of reporter genes such as CAT (3),
luciferase (Fig. 3) (4,12,21), or green fluorescent protein (GFP) (Fig. 4) is
driven by gene regulatory sequences, such as those regulating COL2A1 tran-
scription, may be transfected in immortalized cell lines to study chondrocyte-
specific responses, as described below. Coexpression of wild-type or
dominant-negative mutants of transcription factors, protein kinases, and other
regulatory molecules, mediated by plasmid or adenoviral vectors, may be per-
formed to dissect further the mechanisms involved.
3.2.1. Transient Transfections Using Luciferase Reporter Plasmids
1. Prepare plasmids using the EndoFree Plasmid Maxi Kit, according to the
manufacturer’s instructions (Qiagen), to generate endotoxin-free DNA (see Note 8).
Fig. 2. Effects of bone morphogenic protein (BMP)-2 and BMP-4 on matrix gene
expression and proteoglycan synthesis by T/C-28a2 immortalized chondrocytes. The
T/C-28 cells were passaged into 6-well plates in DMEM/F-12 containing 10% FCS
and incubated at 37°C for 2 d, at which time the medium was changed to serum-free
medium containing 1% Nutridoma. On d 3, BMP-2 (100 ng/mL) or BMP-4 (100 ng/
mL) was added to wells as indicated, and incubations were continued until d 5.
(A) Total RNA was extracted using TRIzol reagent and mRNAs encoding COL2A1,
aggrecan, biglycan, decorin, and reduced glyceraldehyde-phosphate dehydrogenase
(GAPDH) were analyzed by RT-PCR, as described (4). DNA ladders are shown in
the left lanes of each panel. Note that different GAPDH primers were used in the left
and right panels. (B) Proteoglycans were biosynthetically labeled with [35S]sulfate in
the presence of ascorbate (25 µg/mL) during the final 12 h of incubation. The cell
layers (C) and media (M) were extracted separately, as described (11), prior to elec-
trophoresis on SDS-polyacrylamide 4–20% gradient gels and visualization by fluo-
rography. Aggrecan, biglycan, and decorin migrate as indicated on the left of the
panel. Molecular weight standards are indicated to the right of the panel.
46 Goldring
Fig. 3. Regulation of COL2A1 promoter activity by interleukin-1β (IL-1β) in

immortalized human chondrocytes. Luciferase reporter constructs containing
COL2A1 sequences were transfected into C-28/I2 cells using LipofectAMINE+ and
treated with IL-1β in DMEM/F-12 containing 1% Nutridoma for 18 h prior to har-
vest for luciferase assay. The promoter constructs, pGL2B4.0, pGL2B-577/+125,
and pGL2B-131/+125 containing the E309 enhancer region (+2388/+2696 bp) from
intron 1 were compared. Luciferase activity was normalized to the amount of pro-
tein and expressed as relative activity to that of untreated cells transfected with
pGL2B-577/+125E309. Each value is calculated as the mean SD of the results
from three to six wells, and results are representative of at least three experiments.
Note that the empty vector, pGL2-Basic, expressed at 1–2% of the levels of the
COL2A1 promoter constructs and did not respond to IL-1β.
Fig. 4. Transient expression of green fluorescent protein in immortalized human

chondrocytes. The C-20/A4 cells were transfected using LipofectAMINE+ with
pEGFP-N2 at 1 µg/well in 6-well plates and cultured for 24 h in medium containing
10% FCS. Live cultures were visualized by fluorescence microscopy (Nikon Eclipse
TE300). Expression of the green fluorescent protein is driven by the CMV promoter in
this vector (BD Biosciences Clontech).
2. On the day before transfection, seed cells in 6-well tissue culture plates at 2.5–5
× 105 cells/well in DMEM/F-12 containing 10% FCS. (Determine optimal cell
number to ensure that cultures are 50–95% confluent at the time of transfection.)
3. On the day of the transfection, prepare lipid/DNA complexes in serum-free
DMEM/F-12 or Opti-MEM using LipofectAMINE+ or FuGENE 6, according to
the manufacturer’s protocol. Prepare in bulk for multiple transfections. Do not
vortex at any step:
a. LipofectAMINE+: for each well, add 92 µL of serum-free medium to a small
sterile polypropylene tube, add 1 µL of plasmid DNA (maximum of 1 µg; see
Note 9), and tap gently to mix. Add 6 µL of PLUS reagent, mix, and incubate
for 15 min at room temperature. Dilute 4 µL of LipofectAMINE+ reagent
into 100 µL of serum-free medium, mix, and add to each reaction mixture.
Mix and leave at room temperature for an additional 15–30 min at room tem-
perature.
b. FuGENE 6: for each well, add 96 µL of serum-free medium to a small sterile
polypropylene tube, add 3 µL of FuGENE 6 reagent, and tap gently to mix.
Add 1 µL plasmid of DNA (maximum of 1 µg) to the prediluted FuGENE 6
reagent and incubate for 15 min at room temperature.
4. While the lipid/DNA complexes are forming, replace culture medium on cells
with serum-free medium to give a final volume of 1 mL. Add the lipid/DNA
complex mixture dropwise to the well and incubate for 4 h at 37°C.
5. Dilute the transfection medium by adding to the wells an equal volume of DMEM/
F-12 containing 2% Nutridoma-SP (or 20% FCS) and incubate 2 h to overnight.
Add test agent without medium change and incubate for a further 18 h (up to 48
h) (see Note 10).
3.2.2. Cotransfections Using Plasmid Vectors

for Expression of Recombinant Proteins
1. Prepare plasmids as described above in Subheading 3.2.1.
2. Titrate each expression vector and its corresponding empty vector, at amounts
ranging from 10 to 200 ng per well, against a fixed amount of reporter vector.
Equalize the total amount of reporter plus expression plasmid in each well (<1 µg/
well) by adding empty vector and maintain equal volumes (see Note 9).
3. After cotransfection, incubate the cells for 18–24 h to permit expression of recom-
binant protein prior to treatment with test reagents.
3.2.3. Adenovirally Mediated Expression of Recombinant Proteins

1. Infect the 293 producer cell line with adenoviral vector containing the cDNA
encoding the wild-type or mutant protein to be coexpressed and determine the
titer (multiplicity of infection [MOI]) by standard techniques.
2. Incubate immortalized chondrocytes in DMEM/F-12 containing 10% FCS for
18 h following transfection of the reporter construct.
3. Remove medium and wash cells with PBS.
48 Goldring
4. Add 1 mL of serum-free medium containing adenovirus at 1:125 MOI. Incubate

at 37°C for 90 min.
5. Add 1 mL of DMEM/F-12 containing 20% FCS and continue incubation for 18 h.
6. Change medium to fresh DMEM/F-12 containing 10% FCS or 1% Nutridoma-
SP, incubate for 1 h, and treat with test agent for 18 h.
3.2.4. Luciferase Assay

1. Prepare cell lysates by extraction with 200 µL of Passive Lysis Buffer, which
will passively lyse cells without the requirement of a freeze-thaw cycle. Scrape
cells with a policeman and transfer solubilized cells to a 1.5-mL microcentrifuge
tube. Microcentrifuge for 5 min at maximum speed at 4°C, transfer supernatant
to a clean microcentrifuge tube, and store on ice.
2. Determine the protein content using the Coomassie Plus Protein Assay Reagent.
3. Determine luciferase activities using the Luciferase Assay System or equivalent,
according to the manufacturer’s protocol. Mix, manually or automatically, 20 µL
of cell lysate with 100 µL of Luciferase Assay Reagent and read in a luminometer.
Normalize to the amount of protein (or internal control such as β-galactosidase)
and express as relative activity against the empty vector or untreated control.
Perform each treatment in triplicate wells and each experiment at least three times
to ensure reproducibility and significance.
4. The Dual-Luciferase Reporter Assay, in which 20 ng of the pRL-TK Renilla
luciferase control vector is included in the transfections, may be used routinely
or when it is necessary to check the purity of new plasmid preparations or rela-
tive activities of mutant and wild-type constructs.
4. Notes
1. Batches of serum should be tested and selected on the basis of the capacity to
support expression of chondrocyte-specific matrix expression. High capacity to
induce cell proliferation is not necessarily associated with the ability to maintain
phenotype.
2. After prolonged passaging in a monolayer, immortalized chondrocytes may be
redifferentiated by 1–2 wk of culture in alginate beads or in suspension over
agarose or polyHEMA. Note that culture of chondrocytes embedded in agarose is
not discussed here because it is difficult to recover cells for further culture or
other manipulations.
3. Since chondrocytes are strongly adherent to tissue culture plastic, possibly because
of calcium ion-binding glycosaminoglycans in the pericellular matrix, a trypsin-
EDTA solution rather than trypsin alone is used for full release of chondrocytes
from tissue culture plastic during passaging. Immortalized adult articular
chondrocytes adhere more strongly to the culture dish than the more rapidly prolif-
erating immortalized juvenile chondrocytes.
4. The synthetic activities of immortalized chondrocytes in monolayer culture are
inversely related to proliferative activities. Thus, the expression of genes encod-
ing matrix proteins and their deposition into the extracellular matrix decrease
compared with housekeeping and cell growth-associated genes. For experiments,

the growth medium is removed 1–3 d after passaging, when the cultures are
subconfluent, and serum-free medium supplemented with an insulin-containing
serum substitute such as Nutridoma-SP or ITS+ is added, followed immediately
1–24 h later by the addition (without medium change) of the test agent of interest.
Confluent cultures may tolerate serum-free medium containing 0.3% bovine
serum albumin for up to 48 h or longer.
5. The method for culture of chondrocytes in alginate beads has been adapted from
previously published methods (24,25). This method was developed for
nonimmortalized articular chondrocytes that do not proliferate when cultured
in fluid or gel suspension. Immortalized chondrocytes continue to proliferate in
alginate, and rapidly proliferating cell lines cannot be kept in alginate longer than
1 or 2 wk. Since the alginate does not remodel, the clonal growth forms bundles
of cells that cannot expand into the matrix, the cytoplasmic volume is reduced,
and cell death may occur in the middle of the cell mass. Immortalized adult
articular chondrocytes, which tend to grow more slowly, can be kept in alginate
culture for longer periods, as described (4,13).
6. Ascorbate, which is required for synthesis and secretion of proteoglycans and
collagens, is added daily to alginate or other 3D cultures to permit secretion and
deposition of extracellular matrix, particularly when staining techniques are to
be used. Add 25 µg/mL of ascorbate during the final 24–72 h of incubation when
radiolabeling proteoglycans with [35S] sulfate or collagens with [3H]proline for
characterization by sodium dodecyl sulfate polyacrylamide gel electrophoresis.
Do not use ascorbate, which inhibits COL2A1 transcription (M. B. Goldring,
unpublished observation), if analysis of type II collagen mRNA is the endpoint.
7. Culture of immortalized chondrocytes in 3D scaffolds is a useful approach for
tissue engineering applications. The commercially available methods are recom-
mended because of their ease of use. Published methods are available for fabri-
cating collagen sponges (26) and other 3D scaffolds, in which the composition
may be manipulated, for example, by using type II collagen and/or adding gly-
cosaminoglycans and other cartilage-specific matrix components. The biodegrad-
able scaffolds are particularly useful if the cell-seeded scaffolds are to be
implanted in animals. For studies entirely in vitro, in which incubation periods of
more than a few days are required, it is recommended that cultures be performed
in wells that fit the size of the scaffolds. Otherwise, the culture surface of the well
or dish should be coated with a nonadherent substrate or treated in such a way as
to prevent attachment of cells that may migrate out from the sponges. Additional
analytical methods have been described using Gelfoam (27) and BD 3D scaf-
folds (see Website: http://www.bdbiosciences.com/discovery_Labware/Products/
tissue_engineering/).
8. Although chondrocytes are generally less susceptible than monocyte/macro-
phages and other immune cells to endotoxin, it is possible that the transfection
conditions, the proliferative state of the cell lines, or other factors may sensitize
the cells to low concentrations of endotoxin (28). Endotoxin itself induces and
50 Goldring
activates transcription factors that are common to inflammatory responses and

may thus up- or downregulate the promoter of interest, thereby masking the
response to a cytokine or growth factor.
9. The total amount of plasmid to be transfected, including reporter, expression, and
internal control plasmids, should not exceed 1 µg, and the optimal amount for the
culture system should be tested empirically. If variable amounts of expression
vector, for example, are included, the total amount of plasmid in each well should
be equalized by the addition of the empty vector. Wells transfected with appropri-
ate empty vector controls, without and with treatment with test agent, should also
be included. Of the luciferase reporter vectors available, pXP2 and pGL2 provide
reproducible results in immortalized chondrocyte cell lines. Results using pGL3
as the reporter vector are variable depending on the promoter and treatment, and
stable transfection may be required when using highly proliferative immortalized
chondrocyte cell lines or transformed cells such as the chondrosarcoma cell line
SW-1351 (29).
10. The times of incubation following transfection before addition of the test agent
may vary according to the cell line and treatment and should be tested empiri-
cally. Nutridoma-SP or another serum substitute may be used for experiments
with highly proliferative immortalized chondrocyte cell lines for the reasons indi-
cated in Note 4. Note also that test agents should be added without medium
change to avoid induction of pathways by serum growth factors or other medium
constituents.
Acknowledgments
We thank Dr. Elisabeth Morris (Wyeth Research, Cambridge, MA) for pro-
viding recombinant BMP-2 and BMP-4. This work was supported in part by
NIH grants AR45378 and AG22021 and a Biomedical Science Grant from the
Arthritis Foundation.
References
1. Benoit, B., Thenet-Gauci, S., Hoffschir, F., Penformis, P., Demignot, S., and
Adolphe, M. (1995) SV40 large T antigen immortalization of human articular
chondrocytes. In Vitro Cell. Dev. Biol. 31, 174–177.
2.
2 Steimberg, N., Viengchareun, S., Biehlmann, F., et al. (1999) SV40 large T anti-
gen expression driven by col2a1 regulatory sequences immortalizes articular
chondrocytes but does not allow stabilization of type II collagen expression. Exp.
Cell. Res. 249, 248–259.
3.
3 Goldring, M. B., Birkhead, J. R., Suen, L.-F., et al. (1994) Interleukin-1β-modu-
lated gene expression in immortalized human chondrocytes. J. Clin. Invest. 94,
2307–2316.
4.
lar chondrocytes maintain cartilage-specific phenotype and responses to interleukin-
1β. Arthritis Rheum. 43, 2189–2201.
5.
5 Grigolo, B., Roseti, L., Neri, S., et al. (2002) Human articular chondrocytes
immortalized by HPV-16 E6 and E7 genes: maintenance of differentiated phe-
notype under defined culture conditions. Osteoarthritis Cartilage 10, 879–889.
6.
6 Piera-Velazquez, S., Jimenez, S. A., and Stokes, D. (2002) Increased life span
of human osteoarthritic chondrocytes by exogenous expression of telomerase.
Arthritis Rheum. 46, 683–693.
7.
7 Oyajobi, B. O., Frazer, A., Hollander, A. P., et al. (1998) Expression of type X
collagen and matrix calcification in three-dimensional cultures of immortalized
temperature-sensitive chondrocytes derived from adult human articular cartilage.
J. Bone Miner. Res. 13, 432–442.
8.
8 Castagnola, P., Moro, G., Descalzi-Cancedda, F., and Cancedda, R. (1986) Type
X collagen synthesis during in vitro development of chick embryo tibial chondro-
cytes. J. Cell Biol. 102, 2310–2317.
9.
9 Reginato, A. M., Iozzo, R. V., and Jimenez, S. A. (1994) Formation of nodular struc-
tures resembling mature articular cartilage in long-term primary cultures of human
fetal epiphyseal chondrocytes on hydrogel substrate. Arthritis Rheum. 37, 1338–1349.
10. Robbins, J. R. and Goldring, M. B. (1998) Methods for preparation of immortal-
ized human chondrocyte cell lines, in Methods in Molecular Medicine: Tissue
Engineering Methods and Protocols (Morgan, J. R. and Yarmush, M. L., eds.),
11 Kokenyesi, R., Tan, L., Robbins, J. R., and Goldring, M. B. (2000) Proteoglycan
11.
production by immortalized human chondrocyte cell lines cultured under condi-
tions that promote expression of the differentiated phenotype. Arch. Biochem.
Biophys. 383, 79–90.
12.
12 Osaki, M., Tan, L., Choy, B. K., et al. (2003) The TATA-containing core pro-
moter of the type II collagen gene (COL2A1) is the target of interferon-γ-medi-
ated inhibition in human chondrocytes: requirement for Stat1α, Jak1, and Jak2.
Biochem. J. 369, 103–115.
13.
13 Thomas, B., Thirion, S., Humbert, L., et al. (2002) Differentiation regulates
interleukin-1β-induced cyclo-oxygenase-2 in human articular chondrocytes: role
of p38 mitogen-activated kinase. Biochem. J. 362, 367–373.
14.
14 Loeser, R. F., Sadiev, S., Tan, L., and Goldring, M. B. (2000) Integrin expression
by primary and immortalized human chondrocytes: evidence of a differential role
for α1β1 and α2β1 integrins in mediating chondrocyte adhesion to types II and
VI collagen. Osteoarthritis Cartilage 8, 96–105.
15.
15 Attur, M. G., Dave, M., Cipolletta, C., et al. (2000) Reversal of autocrine and para-
crine effects of interleukin 1 (IL-1) in human arthritis by type II IL-1 decoy receptor.
Potential for pharmacological intervention. J. Biol. Chem. 275, 40,307–40,315.
16.
16 Nuttall, M. E., Nadeau, D. P., Fisher, P. W., et al. (2000) Inhibition of caspase-3-
like activity prevents apoptosis while retaining functionality of human
chondrocytes in vitro. J. Orthop. Res. 18, 356–363.
17.
17 Johnson, K., Vaingankar, S., Chen, Y., et al. (1999) Differential mechanisms of
inorganic pyrophosphate production by plasma cell membrane glycoprotein-1 and
B10 in chondrocytes. Arthritis Rheum. 42, 1986–1997.
52 Goldring
18.
18 Palmer, G., Guerne, P. A., Mezin, F., et al. (2002) Production of interleukin-1
receptor antagonist by human articular chondrocytes. Arthritis Res. 4, 226–231.
19.
19 Laflamme, C., Filion, C., Bridge, J. A., Ladanyi, M., Goldring, M. B., and Labelle,
Y. (2003) The homeotic protein Six3 is a coactivator of the nuclear receptor NOR-
1 and a corepressor of the fusion protein EWS/NOR-1 in human extraskeletal
myxoid chondrosarcomas. Cancer Res. 63, 449–454.
20.
20 Grall, F., Gu, X., Tan, L., et al. (2003) Responses to the pro-inflammatory
cytokines interleukin-1 and tumor necrosis factor-α in cells derived from rheuma-
toid synovium and other joint tissues involve NF-κB-mediated induction of the
Ets transcription factor ESE-1. Arthritis Rheum. 48, 1249–1260.
21. Tan, L., Peng, H., Osaki, M., et al. (2003) Egr-1 mediates transcriptional repression
of COL2A1 promoter activity by interleukin-1β. J. Biol. Chem. 278, 17,688–17,770.
22.
22 Ulrich-Vinther, M., Maloney, M. D., Goater, J. J., et al. (2002) Light-activated
gene transduction enhances adeno-associated virus vector-mediated gene expres-
sion in human articular chondrocytes. Arthritis Rheum. 46, 2095–2104.
23.
23 Zwicky, R., Muntener, K., Goldring, M. B., and Baici, A. (2002) Cathepsin B
expression and down-regulation by gene silencing and antisense DNA in human
chondrocytes. Biochem. J. 367, 209–217.
24. Guo, J., Jourdian, G. W., and MacCallum, D. K. (1989) Culture and growth char-
acteristics of chondrocytes encapsulated in alginate beads. Connect. Tiss. Res. 19,
277–297.
25.
25 Hauselmann, H. J., Fernandes, R. J., Mok, S. S., et al. (1994) Phenotypic stability
of bovine articular chondrocytes after long-term culture in alginate beads. J. Cell
Sci. 107, 17–27.
26.
26 Mizuno, S., Allemann, F., and Glowacki, J. (2001) Effects of medium perfusion
on matrix production by bovine chondrocytes in three-dimensional collagen
sponges. J. Biomed. Mater. Res. 56, 368–375.
27.
27 Wu, Q. Q. and Chen, Q. (2000) Mechanoregulation of chondrocyte proliferation,
maturation, and hypertrophy: ion-channel dependent transduction of matrix defor-
mation signals. Exp. Cell Res. 256, 383–391.
28.
28 Cotten, M., Baker, A., Saltik, M., Wagner, E., and Buschle, M. (1994) Lipopoly-
saccharide is a frequent contaminant of plasmid DNA preparations and can be toxic
to primary human cells in the presence of adenovirus. Gene Ther. 1, 239–246.
29. Mengshol, J. A., Vincenti, M. P., and Brinckerhoff, C. E. (2001) IL-1 induces
collagenase-3 (MMP-13) promoter activity in stably transfected chondrocytic
cells: requirement for Runx-2 and activation by p38 MAPK and JNK pathways.
Nucleic Acids Res. 29, 4361–4372.
Stem Cell Isolation and Chondrogenic Induction 53
5
Generation of Pluripotent Stem Cells
and Their Differentiation to the Chondrocytic Phenotype
Luis A. Solchaga, Jean F. Welter,

Donald P. Lennon, and Arnold I. Caplan
Summary
It is well documented that adult cartilage has minimal self-repair ability. Current methods
for treatment of cartilage injury focus on the relief of pain and inflammation and have met with
limited long-term success. In the forefront of new therapeutic approaches, autologous chondro-
cyte transplantation is still only applied to a very small percentage of the patient population.
Our laboratory has focused on cartilage repair using progenitor cells and studied their dif-
ferentiation into cartilage. Adult mesenchymal stem cells are an attractive candidate as pro-
genitor cells for cartilage repair because of their documented osteogenic and chondrogenic
potential, ease of harvest, and ease of expansion in culture; furthermore, their use will obviate
the need for harvesting precious healthy cartilage from a patient to obtain autologous
chondrocytes for transplantation. However, the need to induce chondrogenic differentiation in
the mesenchymal stem cells is superposed on other technical issues associated with cartilage
repair; this adds a level of complexity over using mature chondrocytes.
This chapter focuses on the methods involved in the isolation of human mesenchymal stem
cells and their differentiation along the chondrogenic lineage. Although we have the technol-
ogy to accomplish chondrogenic differentiation of stem cells, much is still to be learned regard-
ing the regulatory mechanisms controlling the lineage transitions and maturation of the
cartilaginous tissue.
Key Words: Mesenchymal stem cell; chondrogenesis; differentiation; tissue culture; trans-
forming growth factor-β; cartilage; chondrocyte; aggregate culture.
1. Introduction
Tissue Engineering has recently emerged as a discipline that combines the
fields of cell biology, engineering, material sciences, and surgery to provide
new functional tissues using living cells, biomatrices, and signaling molecules
53
54 Solchaga et al.
(1–3). In the past decade, this relatively new discipline has greatly expanded,
with numerous research groups focused on the development of strategies for
the repair and regeneration of a variety of tissues (4,5).
Many of these tissue-engineered approaches have targeted the musculoskel-
etal system in general, with special emphasis on articular cartilage (6–14).
Articular cartilage is especially attractive as a target for tissue engineering strat-
egies because it has been well documented that injuries of the articular cartilage,
an avascular tissue without direct access to a significant source of reparative
cells, do not heal spontaneously (15–20). The vast majority of approaches to the
repair or regeneration of articular cartilage are cell-based, aiming to provide a
population of reparative cells to the injured site. Cells used to develop these
strategies can be either differentiated chondrocytes, isolated from unaffected
areas of the joint surface (12,21–32), or progenitor cells, capable of differentiat-
ing into chondrocytes, which can be isolated from a variety of tissues (33–46).
Harvesting a tissue biopsy from valuable healthy articular cartilage that cannot
repair itself does not seem to be a good choice; therefore a number of research
efforts are directed to the isolation of progenitor cells and the understanding of
the mechanisms involved in their chondrogenic differentiation.
The isolation and mitotic expansion of adult human bone marrow-derived
mesenchymal stem cells (47–50), and the specific culture conditions developed
for their chondrogenic differentiation (51–54) are described in this chapter.
2. Materials
1. Scalpel blade (Fisher).
2. 20-, 10-, and 5-mL Syringes (Becton Dickinson).
3. Lidocaine (Henry Schein).
4. 11-Gage Jamshidi needle (Henry Schein).
5. Sterile gauze (Henry Schein).
6. Heparin (400 U/mL; Henry Schein).
7. Tissue culture vessels:
a. 15- and 50-mL Centrifuge tubes (Falcon).
b. 50-mL Polycarbonate capped centrifuge tubes (Fisher).
c. 15-mL Polypropylene centrifuge tubes (Falcon).
d. Pipets and tissue culture dishes and flasks (Falcon).
8. Tissue culture media:
a. Dulbecco’s modified Eagle’s medium, low glucose (1.5 g/L; DMEM-LG;
Sigma).
b. DMEM, high glucose (4.5 g/L; DMEM-HG; Sigma).
9. Culture media supplements:
a. Fetal bovine serum (FBS; best available).
b. Bovine calf serum (BCS) (Hyclone).
c. ITS+ Premix Tissue Culture Supplement (Becton Dickinson).

d. Stock solution of 10–5 M dexamethasone in DMEM-HG (100X). Dexametha-
sone (Sigma) is dissolved at 10–3 M in absolute ethanol and diluted 1:100 with
DMEM-HG to the 10–5 M stock solution.
e. Stock solution of 100 µM ascorbate-2-phosphate (Wako USA) in H2O (100X).
f. MEM sodium pyruvate (100X; Gibco).
g. Stock solution of 1 µg/mL transforming growth factor-β1 (TGF-β1; R&D
systems) in 4 mM hydrochloric acid (HCl) and 1% bovine serum albumin
(100X).
10. 4% Acetic acid (Fisher) in H2O.
11. 70% Ethanol in H2O.
12. 15% Sodium hypochlorite (bleach) in H2O.
13. Tyrode’s salt solution (Sigma).
14. 0.25% Trypsin 53 mM ethylenediamine tetraacetic acid (EDTA) (Gibco-BRL).
15. Percoll gradient: 22.05 mL Percoll (Sigma), 2.45 mL 1.5 M NaCl, 10.5 mL Tyrode’s
salt solution.
a. Prepare the Percoll solution in any convenient multiples of these volumes and
mix thoroughly.
b. Add 35 mL of Percoll solution per 50-mL polycarbonate capped sterile cen-
trifuge tubes.
c. Spin at 21,200g for 15 min in an SS-34 fixed-angle rotor using a Sorvall cen-
trifuge. The resulting density gradient equals 1.03–1.12 g/L.
d. Percoll can be stored for short periods at 4°C.
16. Benchtop centrifuge.
17. Class II biological safety cabinet.
3. Methods
3.1. Bone Marrow Aspiration
In our Institution, bone marrow is harvested from the posterior superior iliac
crest by physicians of the Department of Hematology-Oncology at University
Hospitals of Cleveland, affiliated with Case Western Reserve University. The
bone marrow sample arrives at our laboratory in a 20-mL syringe and is pro-
cessed as described in the following steps. The details of the bone marrow
aspiration were provided by Dr. Omer Koc of the Department of Hematology-
Oncology and are presented here for the reader’s information. Research proto-
cols involving human bone marrow must be approved by the Institutional
Review Board of the hospital, and donors must give informed consent.
1. Marrow donors who have given informed consent lie in a lateral decubitus position.
2. Locate the posterior superior iliac crest.
3. Wipe the donor’s skin over the superior iliac crest with Betadine.
4. Anesthetize with lidocaine (1%) the skin and subcutaneous tissue superficial to
the iliac crest.
56 Solchaga et al.
Fig. 1. Bone marrow aspiration from the superior posterior iliac crest of a volunteer.
5. Make a small incision (0.5 cm) through the skin and subcutaneous tissue with a
scalpel.
6. Insert an 11-gage Jamshidi needle through the cut and anchor it into the posterior
superior iliac crest. Once the needle is anchored, remove the hub.
7. Attach a 10- or 20-mL syringe containing 2 mL of heparin (preservative-free,
400 U/mL) to the needle.
8. Aspirate the marrow by pulling the syringe plunger back briskly (Fig. 1).
9. Rotate the needle 90° clockwise several times and aspirate the marrow at each
new position.
10. Remove the needle from the iliac crest and apply pressure to the skin, until the
bleeding stops.
11. Transfer the sample to a class II biological safety cabinet (see Note 1).
3.2. Isolation and Seeding of Human Mesenchymal Stem Cells

Human mesenchymal stem cells (MSCs) are isolated by their ability to adhere
to tissue culture plastic and proliferate under specific culture conditions (55).
The original cell suspension seeded on the culture dishes contains a mixture of
cells, including red blood cells, nucleated cells of the hematopoietic lineage,
monocytes, macrophages, and fibroblast-like cells from the bone marrow stroma.
Erythrocytes and leukocytes do not attach to the tissue culture surface and are
eventually removed during the medium changes.
Following is the procedure for MSC isolation and seeding.
1. Eject the contents of the syringe into a sterile 50-mL disposable centrifuge tube
(Fig. 2). Add 25 mL of complete medium (DMEM-LG supplemented with 10%
FBS).
Fig. 2. Transfer of the bone marrow aspirate to a 50-mL centrifuge tube.
2. Pipet up and down to mix the medium and bone marrow sample thoroughly and
transfer a small aliquot (0.2 mL) to a 15-mL centrifuge tube for a preliminary cell
count.
3. Centrifuge the rest of the sample at 500g for 5 min on a benchtop centrifuge.
4. While the sample is being centrifuged, determine the number of cells in the ali-
quot, as follows:
a. Transfer 50 µL of the sample from step 2 to a 0.5- or 1.5-mL microcentrifuge
tube.
b. Add 50 µL of serum-containing medium.
c. Add 100 µL 4% acetic acid and mix in order to lyse the erythrocytes.
d. Count the nucleated cells with a hemacytometer and determine the total num-
ber of nucleated cells in the 50-mL tube.
5. After the sample has been centrifuged, remove the fat layer and supernatant,
aspirating carefully from the top, and being careful not to disturb the pellet.
6. Determine the number of tubes of preformed Percoll (density 1.03–1.12 g/mL)
required to fractionate the nucleated cells based on the preliminary cell count
(49). The guideline is to load no more than 200 × 106 cells per tube. Adjust the
volume of the pellet to allow 5 mL of cell suspension per tube of Percoll. In some
cases, the volume of the pellet will exceed 5 mL even though the cell number is
below 200 × 106. In that case, divide the volume between two tubes of Percoll.
7. Load the proper volume of cell suspension carefully onto the top of a Percoll
gradient. Transfer the suspension slowly so that cells will remain at the top of the
gradient (Fig. 3).
8. Carefully transfer the tubes to a centrifuge and spin in an SS-34 rotor in a Sorvall
centrifuge at 400g for 15 min with the brake off.
58 Solchaga et al.
Fig. 3. Bone marrow sample being layered over the preformed Percoll gradient.
Fig. 4. Sample after gradient fractionation; note the red blood cells at the bottom of
the tube and a band of nucleated cells at the top of the gradient.
9. Return the sample (Fig. 4) to the biological safety cabinet, carefully pipet off the
top layer or layers of cells (about 10–14 mL); (Fig. 5), and transfer this volume to
another sterile 50-mL centrifuge tube (see Note 2).
10. Add serum-containing medium to a volume of 50 mL.
11. Spin in a benchtop centrifuge at 500g for 5 min.
12. Remove and discard supernatant and resuspend the cell pellet in 10 mL of com-
plete medium.
Fig. 5. Recovery of cells from the top fraction of the gradient.
Fig. 6. Inoculation of tissue culture dishes with the cell suspension.
13. Use a 1-mL pipet to take a small sample (about 100 µL) for determination of cell
number as described in step 4.
14. Plate cells at 1.7 × 106 per cm2 in tissue culture dishes or flasks in complete
medium (see Note 3 and Fig. 6).
3.3. Mesenchymal Stem Cell Culture
MSCs are cultured at 37°C in a humidified atmosphere of 95% air and 5%
CO2. Only a small percentage of the cells in the original cell suspension attach
60 Solchaga et al.
Fig. 7. Microscopic appearance of a colony in a primary culture of MSCs. Original

magnification ×4.
to the tissue culture surface. These fibroblast-like cells proliferate and form
loose colonies of spindle-shaped cells that are usually visible under the micro-
scope between d 4 and 6 of culture. These colonies increase in size over the
next 7–10 d, and should be subcultured before the cells become confluent. (It is
important to note that proliferation of these cells is not contact-inhibited.) No
efforts are made to remove nonadherent cells. Medium is pipeted or aspirated
from the dish without vigorous swirling or rinsing; it is hypothesized that these
cells provide cytokines and other factors needed for the growth of the attached
cells. Change the medium twice a week.
1. Aspirate spent medium from plates.
2. Add fresh complete medium.
3.4. Mesenchymal Stem Cell Subculture

MSCs must be subcultured before they reach confluence. In primary cul-
tures, the main consideration to determine when the cells should be subcultured
is cell density within single colonies rather than the overall level of confluence
of the plate, which in the case of primary culture may be low, depending on the
colony-forming efficiency of the population. Typically, MSCs are passaged
when colonies are 80–90% confluent (Fig. 7).
Primary cultures are usually subcultured around d 14 of culture ( 3 d).
1. Aspirate expired medium from plates.
2. Rinse the cell layer with the appropriate volume (5 mL for a 56-cm2 tissue cul-
ture dish) of Tyrode’s salt solution to wash cells.
3. Repeat rinse.
4. Add the appropriate volume of trypsin-EDTA (4 mL for a 56-cm2 tissue culture
dish).
5. Return to the incubator for 5–7 min. Keep the time of exposure as brief as pos-
sible.
6. When most of the cells have become well rounded or have detached from the
tissue culture surface, stop the reaction by adding bovine calf serum (BCS) equal
to half the volume of trypsin-EDTA.
7. Draw up the cell suspension with a pipet and, with the same pipet, use the sus-
pension to wash the remaining cells gently from the dish. It is not necessary (or
desirable) to remove all the cells from the dish, as most of the nonfibroblastoid
cells (which are not likely to be MSCs) in these cultures are more trypsin-resis-
tant than the spindle-shaped fibroblast-like cells. Thus, trypsinization represents,
after attachment of fibroblastic cells to plastic, the second component of the pro-
cess of selection of MSCs from the total marrow cell population.
8. Transfer the cell suspension from all the cultures to an appropriate size centri-
fuge tube or tubes.
9. Spin in benchtop centrifuge at 500g for 5 min.
10. Resuspend pellet in 5–10 mL of complete medium.
11. Determine cell number with a hemocytometer.
12. Plate cells at 3.5–4.0 × 103 per cm2.
Further subculturing is conducted according to the above protocol except
for the following considerations:
1. Subcultured hMSCs are evenly distributed on the tissue culture vessel surface
(Fig. 8) and are not in colonies as for primary cultures. Therefore, the degree of
confluence determines when the cells should be trypsinized; as a general rule,
hMSCs should be trypsinized before they become confluent.
2. Passaged hMSCs are more easily trypsinized than are primary cultures, so expo-
sure of these cells to trypsin is usually limited to 5 min.
MSCs have a high proliferative capacity and may be subcultured multiple
times. During the process of cell expansion, MSCs maintained in complete
medium remain in an undifferentiated state (56).
Subcultured MSCs remain spindle-shaped fibroblastic cells, distributed
evenly over the culture dish. They are also slightly larger than primary cells, a
feature that becomes more pronounced with additional subcultivation.
3.5. Chondrogenic Induction
Bone marrow-derived MSCs can be induced to differentiate into
chondrocytes under specific culture conditions. These conditions include 3D
conformation of the cells in aggregates in which high cell density and cell–
cell interaction play an important role in the mechanism of chondrogenesis.
Together with these physical culture conditions, a defined culture medium is
required to achieve chondrogenic differentiation (51,52,57).
62 Solchaga et al.
Fig. 8. Microscopic appearance of MSCs in first passage. Original magnification ×4.
1. Centrifuge passaged cells in a benchtop centrifuge at 500g for 5 min.

2. Resuspend cells at a density of 5 × 105 cells/mL in DMEM-HG supplemented
with 1% ITS+ Premix, 10–7 dexamethasone, 1 µM ascorbate-2-phosphate, 1%
MEM sodium pyruvate, and 10 ng/mL TGF-β1.
3. Transfer 0.5-mL aliquots of the cell suspension (2.5 × 105 cells) into 15-mL
polypropylene centrifuge tubes.
4. Spin aliquots in a benchtop centrifuge at 500g for 5 min.
5. Loosen the caps of the tubes to allow air exchange (see Note 4).
6. Place the tubes (with loose caps) in the incubator at 37°C in a humidified atmo-
sphere of 95% air and 5% CO2 for up to 3 wk.
7. The cells form an aggregate at the bottom of the tubes 24 h after centrifugation.
8. The tubes must be gently tapped to ensure that the cell aggregate releases from
the walls of the tube and moves freely within the volume of medium.
Change the chondrogenic medium every other day.
1. Carefully aspirate the expired medium using a sterile Pasteur pipet.
2. Add a 0.5-mL aliquot of fresh chondrogenic medium to the tube.
Under these culture conditions, human MSCs undergo chondrogenic differ-
entiation within 2–3 wk, producing abundant extracellular matrix composed
primarily of cartilage-specific molecules such as type II collagen and aggrecan
(Fig. 9).
The initial cell aggregates contain type I collagen and no cartilage-specific
molecules. Type II collagen is typically detected in the matrix by d 5. By d 14,
type II and type X collagen are detected throughout the cell aggregates, except
for an outer layer of flattened cells that remains rich in type I collagen.
Fig. 9. MSC aggregate after 3 wk in chondrogenic conditions.
Aggrecan and link protein can also be detected in extracts of the aggregates
(51,52).
4. Notes
1. The bone marrow sample and all cells derived from it are treated with universal
precautions. Appropriate waste containers must be used for all sharp and nonsharp
disposable supplies that come into contact with human tissue or cells, as well as
the spent culture medium of these cells. The marrow sample is processed in a
class II biological safety cabinet. Personnel processing these samples must wear
proper personal protective equipment, including a lab coat, goggles or a face
shield, gloves, and a surgical mask.
2. Most of the protocols described in this chapter are routine tissue culture proto-
cols with a very low level of difficulty. Recovering the cells from the density
gradient may be one of the steps that requires some practice. Usually the mono-
nucleated cell layer remains at the top of the gradient; the best way to collect
them is to keep the tip of the pipet just above the surface of the solution as if we
were vacuuming the cells from the surface of the liquid. To ensure a good recov-
ery of the mononucleated cells, approximately one-third of the volume of the
density gradient is collected in this manner.
3. Probably the single most important technical tip is the selection of the appropriate
serum lot that supports the proliferation and maintenance of multipotentiality (58)
of bone marrow-derived MSCs. A detailed description of this screening process
can be found in Lennon et al. (1996) (55). Briefly, MSC preparations are divided
in several groups of plates and grown in media supplemented with 10% FBS from
a battery of serum samples. The cell morphology, colony morphology, colony-
64 Solchaga et al.
forming efficiency, and proliferation rate of the cells grown under the different
conditions are recorded through primary, first, and second passage cultures. At
the end of the second passage, once the sufficient cell number has been obtained,
the cells are assayed for their multipotentiality both in vitro (osteogenesis, chon-
drogenesis, and adipogenesis) and in vivo (osteogenesis and chondrogenesis) (58).
The selection of a particular serum lot is based on the results of this battery of
different assays and is, therefore, complicated. Detailed protocols for these assays
and for the measurement of the outcome parameters that define the
mulipotentiality of MSCs can be found in the chapter “Mesenchymal stem cells”
by Lennon and Caplan in the book Culture of Cells for Tissue Engineering edited
by Freshney and Vunjak currently in preparation.
4. Once the cells are introduced into chondrogenic conditions, it is very important
to ensure that the caps of the tubes are loose, so that the air in the tubes can
equilibrate with the 5% CO2, humidified air of the incubator. If the caps are too
tight there will be no air exchange, and the cells will probably die owing to lack
of oxygen and/or reduced pH.
References
1.
1 Langer, R. and Vacanti, J. P. (1993) Tissue engineering. Science 260, 920–926.
2.
2 Vacanti, C. A. and Vacanti, J. P. (2000) The science of tissue engineering. Orthop.
Clin. North Am. 31, 351–356.
3.
3 Vacanti, J. and Langer, R. (1999) Tissue engineering: the design and fabrication
of living replacement devices for surgical re-construction and transplantation.
Lancet 354(Suppl. 1), SI32–SI34.
4. Bonassar, L. J. and Vacanti, C. A. (1998) Tissue engineering: the first decade and
beyond. J. Cell Biochem. Suppl. 30–31, 297–303.
5.
5 Pearson, R. G., Bhandari, R., Quirk, R. A., and Shakesheff, K. M. (2002) Recent
advances in tissue engineering: an invited review. J. Long Term Eff. Med. Implants
12, 1–33.
6.
6 Temenoff, J. S. and Mikos, A. G. (2000) Review: tissue engineering for regenera-
tion of articular cartilage. Biomaterials 21, 431–440.
7. O’Driscoll, S. W. (2001) Preclinical cartilage repair: current status and future
perspectives. Clin. Orthop. 391(Suppl.), S397–S401.
8.
8 Luyten, F. P., Dell’Accio, F., and De Bari, C. (2001) Skeletal tissue engineering:
opportunities and challenges. Best Pract. Res. Clin. Rheumatol. 15, 759–769.
9.
9 Musgrave, D. S., Fu, F. H., and Huard, J. (2002) Gene therapy and tissue engi-
neering in orthopaedic surgery. J. Am. Acad. Orthop. Surg. 10, 6–15.
10.
10 Risbud, M. (2001) Tissue engineering: implications in the treatment of organ and
tissue defects. Biogerontology 2, 117–125.
11.
11 Guilak, F., Butler, D. L., and Goldstein, S. A. (2001) Functional tissue engineering:
the role of biomechanics in articular cartilage repair. Clin. Orthop. 391(Suppl.),
S295–S305.
12.
12 Risbud, M. V. and Sittinger, M. (2002) Tissue engineering: advances in in vitro
cartilage generation. Trends Biotechnol. 20, 351–356.
13.
13 Caplan, A. I. and Goldberg, V. M. (1999) Principles of tissue engineered regen-
eration of skeletal tissues. Clin. Orthop. 367(Suppl.), S12–S16.
14.
14 Caplan, A. I., Elyaderani, M., Mochizuki, Y., Wakitani, S., and Goldberg, V. M.
(1997) Principles of cartilage repair and regeneration. Clin. Orthop. Rel. Res. 342,
254–269.
15.
15 Buckwalter, J. (1998) Articular cartilage: injuries and potential for healing. J.
Orthop. Sports Phys. Ther. 28, 192–202.
16 Buckwalter, J. and Mankin, H. (1998) Articular cartilage repair and transplanta-
16.
tion. Arthritis Rheum. 41, 1331–1342.
17.
17 Buckwalter, J. A. and Mankin, H. J. (1998) Articular cartilage: degeneration and
osteoarthritis, repair, regeneration, and transplantation. Instr. Course Lect. 47,
487–504.
18.
18 Mankin, H. J. (1974) The reaction of articular cartilage to injury and osteoarthritis
(first of two parts) N. Engl. J. Med. 291, 1285–1292.
19.
19 Mankin, H. J. (1974) The reaction of articular cartilage to injury and osteoarthritis
(second of two parts) N. Engl. J. Med. 291, 1335–1340.
20 Mankin, H. J. and Buckwalter, J. A. (1996) Restoration of the osteoarthrotic joint
20.
[editorial]. J. Bone Joint Surg. Am. 78, 1–2.
21.
21 Chesterman, P. J. and Smith, A. U. (1968) Homotransplantation of articular carti-
lage and isolated chondrocytes. An experimental study in rabbits. J. Bone Joint
Surg. Br. 50, 184–197.
22.
22 Bentley, G. and Greer, R. B. d. (1971) Homotransplantation of isolated epiphy-
seal and articular cartilage chondrocytes into joint surfaces of rabbits. Nature 230,
385–388.
23.
23 Wakitani, S., Kimura, T., Hirooka, A., et al. (1989) Repair of rabbit articular sur-
faces with allograft chondrocytes embedded in collagen gel. J. Bone Joint Surg.
Br. 71, 74–80.
24 Vacanti, C. A., Kim, W., Schloo, B., Upton, J., and Vacanti, J. P. (1994) Joint
24.
resurfacing with cartilage grown in situ from cell-polymer structures. Am. J. Sports
Med. 22, 485–488.
25. Freed, L. E., Grande, D. A., Lingbin, Z., Emmanual, J., Marquis, J. C., and Langer,
R. (1994) Joint resurfacing using allograft chondrocytes and synthetic biodegrad-
able polymer scaffolds. J. Biomed. Mater. Res. 28, 891–899.
26.
26 Frenkel, S. R., Toolan, B., Menche, D., Pitman, M. I., and Pachence, J. M. (1997)
Chondrocyte transplantation using a collagen bilayer matrix for cartilage repair.
J. Bone Joint Surg. Br. 79, 831–836.
27.
27 Nehrer, S., Breinan, H. A., Ramappa, A., et al. (1997) Canine chondrocytes seeded
in type I and type II collagen implants investigated in vitro. [erratum appears in
(1997) J. Biomed. Mater. Res. 38(4), 288]. J. Biomed. Mater. Res. 38, 95–104.
28
28. Breinan, H. A., Minas, T., Hsu, H. P., Nehrer, S., Sledge, C. B., and Spector, M.
(1997) Effect of cultured autologous chondrocytes on repair of chondral defects
in a canine model. J. Bone Joint Surg. Am. 79, 1439–1451.
29.
29 Minas, T. (1998) Chondrocyte implantation in the repair of chondral lesions of
the knee: economics and quality of life. Am. J. Orthop. 27, 739–744.
66 Solchaga et al.
30. Giannini, S., Buda, R., Grigolo, B., and Vannini, F. (2001) Autologous chondro-
cyte transplantation in osteochondral lesions of the ankle joint. Foot Ankle Int. 2,
513–517.
31.
31 Breinan, H. A., Minas, T., Hsu, H. P., Nehrer, S., Shortkroff, S., and Spector, M.
(2001) Autologous chondrocyte implantation in a canine model: change in com-
position of reparative tissue with time. J. Orthop. Res. 19, 482–492.
32.
33.
33 Amiel, D., Coutts, R. D., Abel, M., Stewart, W., Harwood, F., and Akeson, W. H.
(1985) Rib perichondrial grafts for the repair of full-thickness articular-cartilage
defects. A morphological and biochemical study in rabbits. J. Bone Joint Surg.
Am. 67, 911–920.
34. O’Driscoll, S. W., Keeley, F. W., and Salter, R. B. (1986) The chondrogenic
potential of free autogenous periosteal grafts for biological resurfacing of major
full-thickness defects in joint surfaces under the influence of continuous passive
motion. An experimental investigation in the rabbit. J. Bone Joint Surg. Am. 68,
1017–1035.
35.
35 Amiel, D., Coutts, R. D., Harwood, F. L., Ishizue, K. K., and Kleiner, J. B. (1988)
The chondrogenesis of rib perichondrial grafts for repair of full thickness articular
cartilage defects in a rabbit model: a one year postoperative assessment. Connect.
Tissue Res. 18, 27–39.
36 Shahgaldi, B. F., Amis, A. A., Heatley, F. W., McDowell, J., and Bentley, G.
36.
(1991) Repair of cartilage lesions using biological implants. A comparative histo-
logical and biomechanical study in goats. J. Bone Joint Surg. Br. 73, 57–64.
37.
37 Wakitani, S., Goto, T., Pineda, S. J., et al. (1994) Mesenchymal cell-based repair
of large, full-thickness defects of articular cartilage. J. Bone Joint Surg. Am. 76,
579–592.
38.
38 Wakitani, S., Goto, T., Pined, S. J., et al. (1994) Mesenchymal cell-based repair
of large, full-thickness defects of articular cartilage. J. Bone Joint Surg. Am. 76,
579–592.
39. Chu, C. R., Coutts, R. D., Yoshioka, M., Harwood, F. L., Monosov, A. Z., and
Amiel, D. (1995) Articular cartilage repair using allogeneic perichondrocyte-
seeded biodegradable porous polylactic acid (PLA): a tissue-engineering study. J.
Biomed. Mater. Res. 29, 1147–1154.
40.
40 Butnariu-Ephrat, M., Robinson, D., Mendes, D. G., Halperin, N., and Nevo, Z.
(1996) Resurfacing of goat articular cartilage by chondrocytes derived from bone
marrow. Clin. Orthop. 330, 234–243.
41.
41 Hunziker, E. B. and Rosenberg, L. C. (1996) Repair of partial-thickness defects in
articular cartilage: cell recruitment from the synovial membrane. J. Bone Joint
Surg. Am. 78, 721–733.
42.
42 Chu, C. R., Dounchis, J. S., Yoshioka, M., Sah, R. L., Coutts, R. D., and Amiel,
D. (1997) Osteochondral repair using perichondrial cells. A 1-year study in rab-
bits. Clin. Orthop. 340, 220–229.
43. Goldberg, V. M., Solchaga, L. A., Lundberg, M., et al. (1999) Mesenchymal stem
cell repair of osteochondral defects of articular cartilage. Semin. Arthroplasty 10,
30–36.
44.
44 Johnstone, B. and Yoo, J. U. (1999) Autologous mesenchymal progenitor cells in
articular cartilage repair. Clin. Orthop. 367(Suppl.), S156–S162.
45.
45 Dounchis, J. S., Bae, W. C., Chen, A. C., Sah, R. L., Coutts, R. D., and Amiel, D.
(2000) Cartilage repair with autogenic perichondrium cell and polylactic acid
grafts. Clin. Orthop. 377, 248–264.
46 Solchaga, L. A., Gao, J., Dennis, J. E., et al. (2002) Treatment of osteochondral
46.
defects with autologous bone marrow in a hyaluronan-based delivery vehicle. Tis-
sue Eng. 8, 333–347..
47.
47 Friedenstein, A. J. (1976) Precursor cells of mechanocytes. Int. Rev. Cytol. 47,
327–359.
48.
48 Caplan, A. I. (1991) Mesenchymal stem cells. J. Orthop. Res. 9, 641–650.
49.
49 Haynesworth, S. E., Goshima, J., Goldberg, V. M., and Caplan, A. I. (1992) Char-
acterization of cells with osteogenic potential from human bone marrow. Bone
13, 81–88.
50 Ashton, B. A., Allen, T. D., Howlett, C. R., Eaglesom, C. C., Hattori, A., and
50.
Owen, M. (1980) Formation of bone and cartilage by marrow stromal cells in
diffusion chambers in vivo. Clin. Orthop. 151, 294–307.
51.
51 Johnstone, B., Hering, T. M., Caplan, A. I., Goldberg, V. M., and Yoo, J. U. (1998)
In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells.
Exp. Cell Res. 238, 265–272.
52.
52 Yoo, J. U., Barthel, T. S., Nishimura, K., et al. (1998) The chondrogenic potential
of human bone-marrow-derived mesenchymal progenitor cells. J. Bone Joint Surg.
Am. 80, 1745–1757.
53.
53 Angele, P., Yoo, J. U. Smith, C., et al. (2000) Cyclic hydrostatic pressure enhances
the chondrogenic phenotype of human mesenchymal progenitor cells differentiated
in vitro. J. Orthop. Res. 21, 451–457.
54 Hanada, K., Solchaga, L. A., Caplan, A. I., et al. (2001) BMP-2 induction and
54.
TGF-β1 modulation of rat periosteal cell chondrogenesis. J. Cell Biochem. 81,
284–294.
55. Lennon, D. P., Haynesworth, S. E., Bruder, S. P., Jaiswal, N., and Caplan, A. I.
(1996) Human and animal mesenchymal progenitor cells from bone marrow: iden-
tification of serum for optimal selection and proliferation. In Vitro Cell. Dev. Biol.
32, 602–611.
56.
56 Jaiswal, N., Haynesworth, S. E., Caplan, A. I., and Bruder, S. P. (1997) Osteo-
genic differentiation of purified culture-expanded human mesenchymal stem cells
in vitro. J. Cell. Biochem. 64, 295–312.
57.
57 Ballock, R. T. and Reddi, A. H. (1994) Thyroxine is the serum factor that regu-
lates morphogenesis of columnar cartilage from isolated chondrocytes in chemi-
cally defined medium. J. Cell Biol. 126, 1311–1318.
58. Pittenger, M. F., Mackay, A. M., Beck, S. C., et al. (1999) Multilineage potential
of adult human mesenchymal stem cells. Science 284, 143–147.
Gene Expression in Chondrocytes 69
6
Semiquantitative Analysis of Gene Expression
in Cultured Chondrocytes by RT-PCR
Gaëlle Rolland-Valognes
Summary
Reverse transcriptase-polymerase chain reaction (RT-PCR) is a powerful, sensitive, and
rapid method to monitor small amounts of nucleic acids. This is of particular interest for small
amounts of cells, as in cartilage. We present here two protocols to isolate total RNA and a
protocol to study matrix metalloproteinase and type II collagen gene expression from
chondrocytes of human origin. Specific gene expression is revealed on an ethidium bromide-
containing agarose gel on an ultraviolet plate and normalized to that of a housekeeping gene.
Key Words: RT-PCR; gene; chondrocytes; cartilage; MMP.
1. Indroduction
Cartilage is composed of specific cells (chondrocytes) embedded in an extra-
cellular matrix primarily composed of collagen (mainly type II), proteoglycans,
and water. In osteoarthritis, this matrix is degraded mainly by enzymes that are
secreted by the chondrocytes themselves, such as the matrix metalloproteinases
(MMPs), regulated at both protein and gene levels.
Because of the relatively small number of cells in cartilage, it is useful to
employ powerful methods to study chondrocyte gene expression. Reverse tran-
scriptase-polymerase chain reaction (RT-PCR) is a powerful, sensitive, and
rapid method to monitor the expression of specific mRNAs.
We first discuss total RNA extraction from human chondrocytes. The expres-
sion of stromelysin-1 (MMP-3), collagenase-3 (MMP-13), and type II collagen
is then used to illustrate the study of gene expression by RT-PCR in human
chondrocytes. Total RNA extracted from tissues or cultured cells is first tran-
scribed in DNA that is called complementary (cDNA). Amplification of cDNA
69
70 Rolland-Valognes
copies of the target mRNA is carried out in a controlled manner, so that the
products of the PCR reaction remain proportional to the amount of mRNA target
in the cell (before the product reaches the plateau). Once deposited on an agar-
ose gel, relative amounts of mRNAs can be studied by densitometric analysis.
2. Materials
2.1. Equipment
1. Centrifuge for 1.5- and 2-mL Eppendorf tubes (Eppendorf 5417R).
2. Mini-Gel Electrophoresis unit (MUPID-2 for DNA, RNA & Proteins, Eurogentec).
3. 0.5-, 1.5- and 2-mL Eppendorf tubes.
4. Filter-tips, presterilized and nucleic acid- and nuclease-free (ART-10, -20, -200
and 1000, Molecular BioProducts).
5. Pipets (Eppendorf P2, P10, P20, P200, and P1000).
6. Spectrophotometer (GeneQuant II, RNA/DNA Calculator, Pharmacia Biotech).
7. Thermocycler GeneAmp PCR System 2400/2700 (Applied Biosystem).
2.2. Cells
1. Human humerus chondrosarcoma cells (SW1353, ATCC HTB-94; Biovalley).
2. Chondrocytes of human origin. For isolation and culture procedures, see Chap-
ters 1, 2, and 14.
2.3. Total RNA Isolation

2.3.1. TRIzol® Method
1. Chloroform (Sigma Aldrich).
2. Isopropyl alcohol (Sigma Aldrich).
3. Diethyl pyrocarbonate-treated water (DEPC-water; 0.1% [v/v] DEPC; Sigma
Aldrich) (see Note 1).
4. 75% Ethanol (in DEPC water).
5. TRIzol reagent (Invitrogen).
2.3.2. RNeasy Mini Kit Method

1. QIAshredder™ spin column (Qiagen).
2. RNeasy® Mini Kit, containing RNeasy Mini spin columns; collections tubes (1.5
and 2 mL); buffers RLT, RW1, and RPE; RNase-free water; and handbook
(Qiagen).
2.3.3. RNA Gels

1. Agarose (Electrophoresis grade, Invitrogen).
2. 3-(N-morpholino)propanesulfonic acid 10× (MOPS; Sigma Aldrich).
3. Formaldehyde, 37% (Sigma Aldrich).
4. RNA sample loading buffer (Sigma Aldrich).
2.4. Reverse Transcription

1. 25 U Oligo-(dT)12–18 (Amersham Biotech).
2. 5000 U RNAGuard (Amersham Biotech).
3. 4 × 25 µmol dNTPs (Eurogentech).
4. Superscript™ II Rnase H– Reverse Transcriptase Kit (200 U/mL), containing
0.1 M dithiothreitol (DTT) and 5X First Strand Buffer (Invitrogen).
2.5. Polymerase Chain Reaction

1. 4 × 25 µmol dNTPs (Eurogentech).
2. 20 µM Primer 3' (Eurogentech).
3. 20 µM Primer 5' (Eurogentech).
4. HotStarTaq DNA polymerase Kit (5 U/mL) containing 10X buffer (Qiagen).
5. Distilled ultrapure H2O.
2.6. Primer Sequences

Gene-specific primers were choosen from corresponding sequences, as shown
in Table 1. Sequence determinations were performed by dideoxy-chain termina-
tion using an automated fluorescent dye ABI 377 DNA sequencer (Applied
Biosystems, France).
2.7. DNA Gels

1. 10 mg/mL Ethidium bromide (Sigma Aldrich).
2. 10X Blue Juice Gel Loading Buffer (Invitrogen).
3. Electrophoresis grade agarose (Invitrogen).
4. 10X Tris-borate EDTA (TBE) electrophoresis buffer (Invitrogen).
3. Methods
The method given here describes extraction and quantification of total RNA
from cells, reverse transcription (RT), polymerase chain reaction (PCR), and
PCR products analysis.
3.1. Extraction and Quantification of Total RNA

When handling RNA, several important precautions must be taken (see
Notes 2–5), as RNA is subject to degradation by intracellular RNases, until it
is flash frozen or homogenized in a lysis buffer.
3.1.1. Extraction Using the RNeasy Mini Kit

A new, rapid, and convenient way of isolating total RNA from cells or tissue
is to use the Qiagen RNeasy Mini Kit, which permits total RNA isolation by
binding to silica gel-based membranes.
72
Table 1
Sequence, Reference, PCR Product Size, and Cycle Number of Primers Used
Gene Sequence Position Size (bp) Cycles Accession No. Ref.
MMP-3 5'-TCA-CTC-ACT-CAC-AGA-CCT-GAC-TCG-3' 763–786 467 25–30 X05232 2

5'-CCT-TAT-CAG-AAA-TGG-CTG-CAT-CG-3' 1207–1229
MMP-13 5'-GGA-ATT-AAG-GAG-CAT-GGC-GAC-3' 506–526 512 30–35 X75308 3
72
5'-TAT-GCA-GCA-TCA-ATA-CGG-TTG-G-3' 996–1017
Pro (α1) II Collagen 5'-AGA-TGG-CTG-GAG-GAT-TTG-AT-3' 702–721 336 35–40 NM_001844 4
5'-CTT-GGA-TAA-CCT-CTG-TGA-CCT-3' 1018–1038
β2M 5'-CTC-GCG-CTA-CTC-TCT-CTT-TCT-GG-3' 1–24 335 20–25 V00567 5
5'-GCT-TAC-ATG-TCT-CGA-TCC-CAC-TTA-A-3' 525–550 J00105
β2M, β2-microglobulin; MMP, matrix metalloproteinase.
Rolland-Valognes
Before starting:
1. Add 10 µL β-mercaptoethanol per mL of buffer RLT.
2. Add 4 volumes of 96–100% ethanol to buffer RPE.
Then
1. Aspirate the culture medium.
2. Lyse cells directly by adding 350 µL (vessel diameter <6 cm) or 600 µL (usual
diameter between 6 and 10 cm) to the cell culture vessel.
3. Pipet the lysate directly onto a QIAshredder Spin column placed in a 2-mL col-
lection tube.
4. Centrifuge for 3 min at maximum speed (20,000g).
5. Add 600 µL of 70° ethanol to the lysate, and homogenize by pipeting.
6. Apply 700 µL of the sample to an RNeasy column.
7. Centrifuge for 15 s at 8000g.
8. Discard the flowthrough material.
9. If the sample exceeds 700 µL, go back to step 6, and centrifuge as above. Discard
the flowthrough material each time.
10. Add 700 µL of buffer RW1 onto the RNeasy column.
11. Centrifuge for 15 s at 8000g.
12. Transfer the column to a new 2-mL collection tube.
13. Add 500 µL of buffer RPE to the column.
14. Centrifuge for 15 s at 8000g, to wash the column.
15. Discard the flowthrough material.
16. Add 500 µL of buffer RPE onto the column.
17. Centrifuge for 2 min at maximum speed (20,000g), to dry the column.
18. Transfer the RNeasy column into a 1.5 mL sterile Eppendorf tube.
19. Pipet 25 µL of Rnase-free water directly onto the membrane.
20. Centrifuge for 1 min at 8000g, to eluate the RNA.
3.1.2. Extraction Using TRIzol®

Alternatively, you can use the protocol derived from Chomczinsky and
Sacchi (1) (see Note 6).
1. Lyse cells directly by adding 1 mL of TRIzol reagent per 10 cm2 culture plate.
2. Incubate the homogenate for 5 min at room temperature (to permit the complete
dissociation of nucleoprotein complexes).
3. Add 0.2 mL of chloroform per tube.
4. Shake tubes vigorously by hand for 15 min.
5. Incubate them for 2–3 min at room temperature.
6. Centrifuge at 12,000g for 15 min at 4°C.
7. RNA is located in the upper-aqueous phase, wich represents about 60% of the
volume of the TRIzol reagent used for homogeneization.
8. Transfer the aqueous phase into a new tube, and precipitate the RNA by mixing
with 0.5 mL of isopropyl alcohol.
9. Incubate the tubes at room temperature for 10 min.
74 Rolland-Valognes
Table 2
Preparation of RNA Agarose Gels
Volume 50.0 mL 100 mL
Agarose 0.5 g 1g
DEPC-H2O 43.5 mL 87 mL
Dissolve agarose by heating for 5–7 min in a microwave oven.
Then, add MOPS buffer and formaldehyde, as follows:
MOPS 10× 5.0 mL 10 mL
Formaldehyde 37% 1.5 mL 3 mL
DEPC, diethyl pyrocarbonate; MOPS, 3-(N-morpholino)propane-

sulfonic acid.
10. Centrifuge at 12,000g for 15 min at 4°C.

11. Wash the RNA pellet once with 75% ethanol.
12. Mix by vortexing.
13. Centrifuge at 7500g for 5 min at 4°C.
14. Let the pellet briefly air-dry (5 min).
15. Redissolve the RNA in a small quantity of DEPC-treated water (20–25 mL).
3.1.3. Quantification
1. Prepare a 1:200 or 1:500 dilution in DEPC-treated H2O for each sample.
2. Determine the RNA quantity and quality by measuring the absorbance at 260 nm
(A260) and 280 nm (A280) for each sample in a spectrophotometer.
3. Calculate the concentration of each sample: 1 U at 260 nm corresponds to 40 µg
of RNA per mL.
4. The A260:A280 ratio gives an estimate of RNA purity (with respect to contami-
nants that absorbs ultraviolet [UV] light, such as proteins). Pure RNA has a ratio
of 1.8–2.0 (see Note 7).
3.2. RNA Gels

1. Prepare a 1% agarose gel in DEPC-H2O, as shown in the Table 2. Pour the gels
into the plates, and allow to cool down for at least 30 min.
2. Prepare the samples, as follows:
a. For each sample, prepare 1.0–2.5 µg of RNA qsp 15 µL DEPC-H2O in a new
tube.
b. Add 3 µL of RNA loading buffer (containing ethiduim bromide [BET])
c. Heat for 5 min at 65°C.
d. Put on ice for 5 min.
e. Put the sample onto the gel.
f. Perform migration for 30 min at 100 V, in 1× MOPS buffer.
g. Look at your gel on a UV plate (see Fig. 1).
Fig. 1. Formaldehyde agarose gel of 2.5 µg of total RNA isolated from human
chondrocytes. Ribosomal bands 28S and 18S are clearly visible. The 28S band should
be twice as big and as intense as the 18S band. Band sizes are given for human RNA.
Table 3
Preparation of RT Mix
Concentration
RT Mix Components Initial Final For 1 tube (µL)
RNAGuard 40 U/mL 40 U 1
First strand buffer 5× 1× 4
dNTPs 10 mM 500 µM 1
DTT 0.1 M 10 mM 2
Superscript II 200 U/µL 200 U 1
Total volume 9
DTT, dithiothreitol.
3.3. Reverse Transcription

Reverse transcribe total RNA (1–2.5 µg) for 1 h at 37°C, using oligo (dT)12–18
and Superscript II, in a total reaction volume of 20 µL according to the
manufacturer’s specification.
1. Prepare 1–2.5 µg of total RNA, qsp 10 µL with DEPC-treated water.
2. Add 1 µL of oligo-(dT)12–18 (500 ng/mL).
3. Incubate for 5 min at 65°C.
4. Place the tubes on ice for 2 min.
5. Prepare RT mix, as indicated in Table 3. (Prepare the mix for one extra sample.)
6. Add 9 µL of mix to each tube and mix smoothly by pipeting. Final volume is 20
µL.
7. Place the tubes into a thermocycler, and start the RT program : 1 h at 37°C, and
then 5 min at 65°C.
8. Store RT products at 4°C or –20°C (for long-term conservation).
3.4. Polymerase Chain Reaction

Perform PCR using 2–4 µL of cDNA in a final volume of 50 µL. The primer
sets used have been chosen from previously published sequences (Table 1). β-2-
Microglobulin is used as a housekeeping gene (see Note 8).
76 Rolland-Valognes
Table 4
Preparation of PCR Mix
Concentration
RT Mix Components Initial Final For 1 tube (µL)
Ultrapure-H2O 39.75
Buffer 10× 1× 5
dNTPs 10 mM 200 µM 1
Primer 3' 20 µM 0.4 µM 1
Primer 5' 20 µM 0.4 µM 1
HotStart Taq
DNA Polymerase 5 U/µL 1.25 U 0.25
Total volume 48
cDNA 1 µg 2
Final volume 50
1. Add 2 µL of RT reaction mix in a PCR tube.

2. Prepare PCR mix, as indicated in Table 4. (Prepare the mix for an extra; see
Note 9.)
3. Add 48 µL of mix to each tube, and mix smoothly by pipeting. The final volume
is 50 µL.
4. Place the tubes into a thermocycler, and start the Hot Start PCR program: 15 min
at 94°C and then 30 at 94°C, 30 s at annealing temperature, and 1 min at 72°C.
Repeat for 20–35 cycles (see Table 1). Then perform the final elongation for
10 min at 72°C. To be sure to be in the linear range, it is useful to prepare several
identical PCR reactions and remove them from the thermocycler at different num-
bers of cycles. Alternatively, you can remove an aliquot of the PCR reaction, by
stopping the thermocycler and restarting it (see Note 10).
5. Store PCR products at 4°C or –20°C (for long-term conservation) (see Note 11).
3.5. PCR Product Analysis

Prepare a 1% agarose gel.
1. Mix 1 g of agarose with 100 mL of 0.5X TBE.
2. Heat in a microwave oven until the agarose is completely dissolved (5–10 min).
3. Let cool down for 2–3 min.
4. Add 1 µL of BET and mix.
5. Pour the liquid in the electrophoresis plate.
6. Allow 30–45 min for the gel to polymerize.
7. Take off the comb.
Fig. 2. Amplification of α1 type II collagen. One microgram of total RNA isolated

from human chondrocytes has been reverse transcribed and amplified. Lanes 1, non-
treated cells (control); lanes 2, cells treated with interleukin-1β (IL-1β) at 10 ng/mL;
lanes 3, cells treated with transforming growth factor-β (TGF-β) at 10 ng/mL. Ten
microliters of the reaction was taken out of each tube at cycles 20, 23, 26, and 30, and
deposited on a 1% gel.
Prepare the samples.

1. Place 10 µL of the PCR reaction in a new tube.
2. Add 3 µL of loading buffer.
3. Mix gently by pipeting.
4. Load 1 µg of molecular weight (mol wt) marker in the first well (1 µL of mol wt
marker + 9 µL TE + 3 µL loading buffer).
5. Load 13 µL of each sample in each well.
6. Allow to migrate for 30 min at 100 V.
7. Once migration is over, put the gel on a UV plate.
8. Photograph the gel.
The number of cycles used for amplification must be within the linear range
for both the housekeeping gene and the gene of interest (Fig. 2). A plot of band
intensity vs cycle number should be linear for both. The intensity of the band
can then be quantified by Image Analysis Software and normalized to that of
the housekeeping gene.
4. Notes
1. DEPC is a strong but not absolute ribonuclease (RNase) inhibitor that inactivates
RNases by covalent modification. Caution: DEPC should be handled with great
care, as it is a suspected carcinogen.
2. RNases are very stable and active enzymes, and they are difficult to inactivate.
Always wear latex or vinyl gloves (as hands and dust particles are a source of
RNAses), and keep tubes closed whenever possible. Use sterile plasticware (gen-
erally RNAse-free) or glassware (which should be cleaned with a detergent,
rinsed, and baked at 240°C for several hours). Alternatively, glassware can be
treated with DEPC. Fill glassware with 0.1% DEPC in water, allow to stand over-
night at room temperature, and then autoclave for 15 min to eliminate residual
DEPC.
78 Rolland-Valognes
3. Solutions or water should also be treated with 0.1% DEPC. Add 0.1 mL DEPC to
100 mL solution and water, shake vigorously to bring DEPC into solution, and
allow the bottle to stay overnight at room temperature. Then autoclave for 15 min
to remove any trace of DEPC.
4. Cell pellets or tissue should be stored at –70°C for several months, for later use.
5. RNA is not protected after harvesting until the cells are lysed in a guanidinium-
based buffer (such as buffer RLT or TRIzol).
6. When using the TRIzol protocol for RNA extraction, you can also isolate DNA
and proteins from the same sample. Please refer to the TRIzol protocol, Sub-
heading 3.1.2.
7. High quality and concentration of the RNA sample will reduce optimization of
the PCR reactions.
8. Housekeeping genes, such as β2-microglobulin or GAPDH are expressed at very
high levels compared with the genes of interest and will require 20–25 cycles for
detection on gels. On the other hand, MMP or collagen mRNAs are expressed at
considerably lower levels and thus require more cycles (25–35). Optimal PCR
cycle numbers have to be determined empirically (as mentioned in the text).
9. Performing HotStart PCR increases the specificity of primer-mediated amplifi-
cation and reduces nonspecific amplifications.
10. If you open PCR tubes during the amplification step, in order to take an aliquot of
the reaction, be careful of crosscontamination owing to aerosols. It is essential to
discard the pipet tips following each operation to prevent carry over contamina-
tion.
11. Keep the final PCR products away from the PCR setup area to reduce the possi-
bilities of crosscontamination. Use aerosol-resistant tips (filtered tips), and use
dedicated pipets for sample preparation and PCR product analysis.
References
1.
1 Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by
acid guanidinium thiocyanate phanol chloroform extraction. Anal. Biochem. 162,
156–159.
2.
2 Whitham, S. E., Murphy, G., Angel, P., et al. (1986) Comparison of human
stromelysin and collagenase by cloning and sequence analysis. Biochem. J. 240(3),
913–916.
3. Freije, J. M., Diez-Itza, I., Balbin, M., eta l. (1994) Molecular cloning and expres-
sion of collagenase-3, a novel human matrix metalloproteinase produced by breast
carcinomas. J. Biol. Chem. 269(24), 16,766–16,773.
4.
4 Cheah, K. S., Stoker, N. G., Griffin, J. R., Grosveld, F. G., and Solomon, E. (1985)
Identification and characterization of the human type II collagen gene (COL2A1).
Proc. Natl. Acad. Sci. USA 82(9), 2555–2559.
5. Suggs, S. V., Wallace, R. B., Hirose, T., Kawashima, E. H., and Itakura, K. (1981)
Use of synthetic oligonucleotides as hybridization probes: isolation of cloned
cDNA sequences for human β2-microglobulin. Proc. Natl. Acad. Sci. USA 78(11),
6613–6617.
mRNA Expression in Articular Chondrocytes 79
7
Quantification of mRNA Expression Levels
in Articular Chondrocytes With PCR Technologies
Audrey McAlinden, Jochen Haag, Brigitte Bau,

Pia M. Gebhard, and Thomas Aigner
Summary
Unlike any other technology in molecular biology, the polymerase chain reaction (PCR) has
changed the technological armamentarium of molecular scientists working on cartilage, in terms
of outstanding sensitivity and accuracy. Four approaches to determine mRNA expression
levels by PCR amplification of specific cDNA sequences are currently in use and are discussed
in this chapter: conventional PCR with end-point determination, conventional PCR in the loga-
rithmic amplification phase, conventional PCR using internal competitive DNA fragments, and
real-time PCR as offered by TaqMan™ technology and others. The determination of mRNA
expression levels by real-time quantitative PCR appears to be the most reliable method for accu-
rate determination of gene expression levels within cartilage and cultured chondrocytes, as in
other tissues and cell types. This technology offers outstanding sensitivity and accuracy in terms
of determination of the amount of cDNA molecules. However, this method cannot account for
factors such as efficiency of RNA isolation and reverse transcription conditions. Thus, normal-
ization of the acquired data is required, with all its limitations as described.
Key Words: PCR; cartilage; chondrocytes; mRNA quantification; expression analysis;

TaqMan.
1. Introduction
Unlike any other technology in molecular biology, the polymerase chain
reaction (PCR) has changed the technological armamentarium of molecular
scientists working on articular cartilage. In addition to the amplification of
DNA fragments for cloning, sequencing, and a large variety of other applica-
tions, the determination of mRNA expression levels of genes in cartilage and
chondrocytes is a widely used PCR application that provides outstanding sen-
sitivity (up to 5–10 molecules per assay) and accuracy (see Subheading 3.4.).
79
80 McAlinden et al.
In general, four approaches to determine mRNA expression levels by PCR

amplification of specific cDNA sequences are currently in use and are discussed
in this chapter: (1) conventional PCR with end-point determination, (2) conven-
tional PCR in the logarithmic amplification phase, (3) conventional PCR using
internal competitive DNA fragments, and (4) real-time PCR as offered by
TaqMan™ technology and others.
2. Materials
1. GeneAmp, PCR System 9700 (Applied Biosystems, Foster City, CA).
2. ABI PRISM® 7700 (Applied Biosystems).
3. RNeasy kit (Qiagen, Hilden, FRG).
4. Random hexamer primers (e.g., Promega, Madison, USA).
5. RNasin (Promega).
6. Moloney murine leukemia virus (MMLV) reverse transcriptase, RNase H Minus,
5X reaction buffer (Promega).
7. Nucleotide mix (dTTP, dCTP, dATP, dGTP; Carl Roth, Karlsruhe, FRG).
8. AmpliTaq Gold® DNA Polymerase, 10X PCR Gold buffer (Applied Biosystems).
9. Silver-Star DNA polymerase; 10X polymerase buffer (Eurogentec, Seraing,
Belgium).
10. qPCR™Core Kit-No Rox; Rox-buffer (Eurogentec).
11. Sachem® LE agarose (BioWhittaker Molecular Applications, Rockland, ME).
12. QIAquick gel extraction kit (Qiagen).
13. Cloning vectors; e.g., pGEM-T Easy (Promega), pCRII TOPO (Invitrogen,
Karlsruhe, FRG).
14. Picogreen® data quantitation kit (Molecular Probes, Eugene, OR).
3. Methods
3.1. Conventional PCR for Detection of mRNA Expression
of SOX9 in Normal and Osteoarthritic Cartilage
3.1.1. Principles
The first step to evaluate the expression of a certain gene is to test for its
presence in a cDNA preparation by conventional PCR. However, it should be
emphasized that conventional PCR is not quantitative, as it mostly determines
the presence or absence of the expression of a certain gene. In addition, levels of
amplified product do not necessarily reflect the total amount of cDNA/mRNA
present in the sample tested owing to the PCR amplification plateau phase. This
effect is largely independent of the amount of starting template. For accurate
detection, more quantitative methods as outlined below should be used or other
mRNA quantification technologies, such as Northern blotting.
The detailed theory of PCR has been described in numerous textbooks (1,2)
and is not the subject of this chapter. Basically, sequence-specific primers induce
the exponential amplification of a region of interest within a template molecule.
3.1.2. Primer Selection

One of the most crucial aspects for designing a PCR is the appropriate selec-
tion of the amplification primers. In this respect some general rules have to be
envisaged, but by using standard computer programs adequate primers can eas-
ily be selected (e.g., Primer Express™, ABI Biosystems):
1. Length of primer: 18–30 bp.
2. The amplification product (as determined by measuring the number of bases
amplified by the 5' forward and 3' reverse primers) should be 150–800 bp (Note:
products smaller than 150 bp may be difficult to detect by conventional agarose
gel electrophoresis, whereas larger products may be more difficult to amplify.)
3. Primers should not display significant complementarity (neither sequence simi-
larity to themselves nor to the other primer).
4. Both forward (5') and reverse (3') primers should have similar melting tempera-
tures (Tm approx 55–70°C; ∆Tm below 1°C).
5. A homology search using public databases and programs (e.g., BLAST, NCBI:
Website: www.ncbi.nlm.nih.gov/blast) allows one to check for potential
crossreaction of the primers to other genes. This problem is usually minimal, as
both primers would need to possess high homology, at least at the 3'-end, to result
in a coamplification product. Crossreactivity may occur, however, if the anneal-
ing and amplification temperature of both primers is not well defined.
3.1.3. Experimental Protocol for Detection of SOX9 mRNA Expression

A typical protocol for an amplification procedure is given below using prim-
ers against human SOX9 (SOX9-fw 5'-primer: 5'-ACCGGCCTCTACTCC
ACCTT-3'; SOX9-dw 3'-primer: 5'-TGGGTACGAGTTGCCTTTAGCT-3').
These primers amplify a product of 331 bp. Conventional PCR showed abun-
dant expression of SOX9 in all cartilage samples investigated, with a lower
signal detected in the osteoarthritic samples (Fig. 1). In fact, quantitative real-
time PCR confirmed this pattern and provided exact values (3).
3.1.3.1. REVERSE TRANSCRIPTION: SYNTHESIS OF CDNA

For all work with RNA, RNAse-free apparatus and material should be used
and gloves should be worn throughout the procedure. The method is described
for components supplied by Promega, but it can be easily adapted for compo-
nents from other suppliers. For RNA preparation from cultured chondrocytes,
see Chapter 6 and for RNA preparation from cartilage, see Chapters 8 and 10.
1. Add 1 µg total RNA into a microtube and 2 µL random hexamers (100 ng/µL; for
poly[dT] primers, see Note 1). Add diethyl pyrocarbonate (DEPC)-treated
(RNAse-free) water to a total volume of 17.25 µL.
2. Incubate for 10 min at 70°C (for denaturation of RNA secondary structures) and
cool immediately on ice for 5 min. Centrifuge briefly.
82 McAlinden et al.
3. Add 5 µL MMLV reverse transcriptase (RT) 5X reaction buffer, 1.25 µL dNTP

mix (10 mM each), 0.5 µL RNasin (40 U/µL), 1 µL MMLV RT (H–) (200 U/µL;
see Note 2).
4. Mix gently and incubate at room temperature for 10 min and then at 37°C for 1 h.
5. Incubate for 15 min at 70°C (to inactivate the enzyme).
3.1.3.2. PCR
1. Prepare a master mix and adjust to a final PCR volume of 20–24 µL with H2O
(depending on the volume of cDNA added): 2.5 µL 10X DNA-polymerase buffer,
2 µL 5'-primer (10 µM), 2 µL 3'-primer (10 µM), 0.25 µL dNTP mix (10 mM
each), 0.75 µL MgCl2 (50 mM), 0.1 µL Taq DNA-polymerase (5 U/µL). Adjust
volume with H2O.
2. Add 1–5 µL of the cDNA reaction (depending on the assay sensitivity).
3. PCR cycle program:
a. Initial denaturation step: 94°C, 4 min (using hot-start polymerase: 10 min).
b. Then 35–40 cycles of: (i) denaturation step: 94°C, 30 s; (ii) annealing step:
approx 60°C (needs to be tested for the respective primers), 30 s; and (iii)
extension step: 72°C, 1 min (followed by a final extension step: 8 min at 72°C).
3.1.3.3. AMPLIFICATION PRODUCT BY AGAROSE GEL ELECTROPHORESIS EVALUATION
Electrophorese an aliquot of the amplification product through a conventional
1–1.5% agarose gel (the optimal agarose concentration depends on the expected
size of the amplification product) and compare the resulting cDNA band size
with appropriate DNA standards. Evaluate the results by ultraviolet (UV) detec-
tion (Fig. 1A). In particular, the size of the cDNA band, the presence of addi-
tional amplification products and the overall intensity of the band should be
noted. The intensity of the band gives some indication of the gene expression
level within the specimen investigated.
Fig. 1. (opposite page) (A) Conventional PCR analysis showing the presence of
SOX9 in normal (lanes 2–4) and osteoarthritic (lanes 5–7) articular cartilage. Amplifi-
cation product is 331 bp. The SOX9 signals are somewhat less intense in the osteoar-
thritic specimens. Lanes 1 and 9, 100 bp DNA ladder standards (MBI Fermentas,
FRG). Lane 8, negative control (no cDNA added). (B,C) Comparative analysis of
cDNA from normal (lanes 1–6) and osteoarthritic (lanes 8–13) cartilage during the
logarithmic phase of the PCR amplification (lanes 1 and 8, 20 cycles; lanes 2 and 9, 23
cycles; lanes 3 and 10, 26 cycles; lanes 4 and 11, 29 cycles; lanes 5 and 12, 32 cycles;
lanes 6 and 13, 35 cycles). Lane 7, negative control. (C) Intensities of the amplifica-
tion bands determined by gel densitometry. Note that at low cycle numbers (20 and
23×), no amplification product is visible, whereas after a high cycle number (35×)
both PCRs have reached the plateau phase, revealing a similar amount of amplified
product, despite differences in starting concentrations of both samples. Quantification
analysis is only possible during the logarithmic amplification phase. OA, osteoarthritic.
84 McAlinden et al.
3.2. Semiquantitative PCR in the Logarithmic Amplification Phase

to Measure SOX9 Expression in Articular Chondrocytes
A more quantitative method to evaluate the amount of cDNA present in a
sample is to titrate the number of amplification cycles in order to detect the
product during the exponential amplification phase of the PCR. This is demon-
strated in Fig. 1B and C: SOX9 was amplified using the same primers and
experimental procedures as described in Subheading 3.1.3. Only the amplifi-
cation cycle numbers differed between experiments. In normal cartilage, more
amplified product is shown at lower cycle numbers compared with that ampli-
fied from osteoarthritic cartilage (compare cycle number 26 and 29 between
normal and osteoarthritic samples). This method should be considered
semiquantitative, as it only allows an approximate estimation of gross differ-
ences in expression levels. Note that the intensities of the product bands at the
plateau phase are identical.
3.3. Semiquantitative PCR Measurement

of Anabolic Chondrocyte Activity Using Competitor Fragments
A more exact, but time-consuming technology for quantifying mRNA
expression levels in chondrocytes is the use of DNA competitor fragments.
Basically, these competitors are similar to the cDNA to be amplified (i.e., they
contain identical primer recognition sites and related internal sequences), thus
ensuring identical amplification efficiency (4). In order to distinguish the
amplified competitor bands from the cDNA amplification products, the inter-
nal competitor sequence should contain either a short deletion or insertion of
additional sequences or a restriction site sequence. This enables one to distin-
guish both products by gel electrophoresis directly, or after digestion with the
specific restriction enzyme.
The experimental concept is based on the addition of a known number of
DNA competitor fragments to the amplification reaction in order to coamplify
them in a competitive manner with the cDNA of interest. The ratio of cDNA/
competitor of the PCR products allows us to calculate the amount of cDNA
originally present in the sample. The basic protocol is very similar to the stan-
dard PCR approach described in Subheading 3.1. After reverse transcription
of RNA, one adds to a series of aliquots of the obtained cDNA a defined
number of competitor fragment molecules and subsequently compares both
amplification products (fragment of interest and competitor fragments) by
conventional gel electrophoresis and densitometry. The number of competitor
fragments that results in a similar amount of amplification product represents
a good estimate for the amount of cDNA molecules present in the starting
material.
3.3.1. Experimental Protocol

The selection of primers should be followed as outlined above in Subheading
3.1.2 (see also Table 1). Care should be taken to optimize the reaction condi-
tions, such as the annealing temperature, primer concentration, nucleotide con-
centration, and optimal number of cycles to be used in order to maximize
visualization and quantitation of the cDNA bands. Different amounts of the com-
petitor fragments should be added at twofold dilution steps over two orders of
magnitude depending on the abundance of cDNA fragments present within the
sample to be measured. These conditions have to be established empirically.
Generation of internal competitor fragments can be performed according to
a protocol originally published by Celi et al. (5), and this is shown in Fig. 2. In
brief, the competitor fragments are shortened versions of the original sequence
of interest (i.e., collagen type II or aggrecan), which are synthesized using spe-
cifically designed internal primers. The competitor fragment generated for col-
lagen type II is 587 bp (instead of 670 bp) and 377 bp for aggrecan (instead of
533 bp). This size difference allows clear distinction by conventional gel elec-
trophoresis. A typical experimental result using these competitors is shown in
Fig 2C.
3.4. Determination of Collagenase mRNA Expression
in Articular Chondrocytes by Real-Time Quantitative PCR
The determination of mRNA expression levels by real-time quantitative
PCR appears to be the most reliable method for accurate determination of gene
expression levels within tissues and cells. This technology offers outstanding
sensitivity and accuracy in terms of determination of the amount of cDNA
molecules. However, this method cannot account for factors such as efficiency
of RNA isolation and reverse transcription conditions. Thus, normalization of
the acquired data is required with all its limitations, as described below (see
Note 3 and Subheading 3.5.). Also, as with all mRNA detection methods, the
ability to deduce protein levels from mRNA data is very limited (see Note 4).
The following section highlights the principles of the TaqMan technology.
(For more detailed information, the reader is referred to the manufacturer’s
manual.) For other real-time PCR technologies, see Note 5.
3.4.1. Principles
The TaqMan assay is a conventional PCR that uses the 5'-exonuclease activ-
ity of Taq DNA polymerase to digest a gene-specific probe (Fig. 3). The deg-
radation of this probe is measured online (i.e., in “real-time”) after each PCR
amplification cycle (Fig. 4A). The probe represents a third oligonucleotide
that contains at its 5'-end a reporter and at its 3'-end a quencher fluorochrome,
as well as a phosphate group to inhibit further elongation during the PCR.
86
Table 1
Sequence of Primers Used for Competitive Polymerase Chain Reaction (PCR)
Experiments in Order to Evaluate Chondrocyte Anabolic Activity a
Tm Ta Amplificate
Target-cDNA Primer Sequence (°C) (°C) (bp)
COL2A1 col2a1-up 5'-CTC-GTC-GCC-GCT-GTC-CTT-CG-3' 67.1 63.0 670

col2a1-low 5'-TCA-CAC-CAG-GAG-CAC-CCG-CC-3' 67.5
86
col2a1-int 5'-CGC-CGC-TGT-CCT-TCG-AAG-GAC-CTC-CTG-GGC-C-3' 48.8 582

aggrecan agg-up 5'-TCC-CTC-ACC-ATC-CCC-TGC-TA-3' 60.5 62.0 533
agg-low 5'-TCT-CCA-TAG-CAG-CCT-TCC-CG-3' 60.6
agg-int 5'-TCC-CTC-ACC-ATC-CCC-TGC-TAC-CTG-CCC-AAC-TAC-CCG-G-3' 56.7 377
aT , theoretical melting temperature (in case of the internal primers, the melting temperature of the sequence specific inner portion); T , optimized annealing
m a
temperature. Bold sequence of the internal primers (“...-int”) represents the portions hybridizing to the internal sequence of the cDNA, whereas the remaining
sequence represents the binding sites of the external primers (or part of it).
McAlinden et al.
Fig. 2. Schematic representation of the synthesis of the internal competitor fragments

for quantitative polymerase chain reaction (PCR) analysis. After normal amplification of
the target fragment from the cDNA template (A), the target fragment is then reamplified
with an internal primer (primer CB), which replaces one of the external primers (B). This
internal primer contains part of the target sequence and usually the sequence of the
replaced external primer. The amplification product represents the shortened competitor
fragment, which is finally added to the PCR. (C) A typical result of a competitive PCR.
Gel electrophoretic analysis shows two amplification products (lower bands represent
the shortened competitor fragment; upper bands represent the original amplified tem-
plates). Three different concentrations of the added competitor fragments were assayed
in 10× dilution steps (artificial numbers: 100, 10, and 1). The concentration of cDNA is
shown to correspond approximately to the intermediate concentration (i.e., 10).
As long as the reporter and the quencher are closely linked by the oligonucle-
otide backbone, no fluorescence signal is detected. However, after degradation
of the oligonucleotide, the reporter is separated from the quencher and can then
be detected by the integrated fluorescence detection system. Importantly, only
probes that are fully hybridized to the template are digested by the Taq DNA
polymerase, whereas free, unhybridized probe is not digested and therefore
does not impede measurement.
88 McAlinden et al.
Fig. 3. The physical principle of probe detection in the TaqMan technology. This is
based on the FET (fluorescence energy transfer) phenomenon (A). If a fluorochrome
Q (i.e., the quencher) is very close to another fluorochrome R (i.e., the reporter) and if
the absorption spectrum of the former is similar to the emission spectrum of the latter,
then after excitation of R, the emitted electron is adsorbed by Q and an electron corre-
sponding to the emission spectrum of Q is emitted. Despite exciting R, no emitted
electrons of R are detectable and, thus, R is quenched by Q. Because this effect relies
very much on the close proximity of R to Q, it is directly linked to the integrity of the
oligonucleotide backbone of the TaqMan probe. (B–D) During the amplification pro-
cedure the polymerase cleaves the backbone (B,C) and R and Q are released and dif-
fuse off. (D) Thereafter, emission photons of R are detectable.
The quantification is usually done in comparison with intra-assay standard

curves (Fig. 4B) and normalized to a housekeeping gene such as GAPDH (glyc-
eraldehyde phosphate dehydrogenase; see also Note 6 and Subheading 3.5.).
The point of measurement is the CT-value of the probe, i. e. the first amplifica-
tion cycle at which the measured signal, representing hydrolyzed probe, exceeds
a certain threshold level (Fig. 4A). Usually, quantification is done using a cali-
bration curve of standard fragments (see Subheading 3.4.2.4.). Assays are best
performed in triplicate in order to account for pipeting errors (see Note 7).
Another quantification method is the ∆∆CT method (see Note 8), which is not
described in this chapter as it is not commonly used.
3.4.2. Primer and Probe Selection and Synthesis
Primers and probes as well as assay conditions need to be carefully selected,
empirically optimized, and tested for the TaqMan assay in order to obtain opti-
mal results.
Fig. 4. (A) On-line measurement of detection of different probes with different

starting concentrations of template to be amplified (curves skewed to the left represent
measurements with higher original concentration and therefore, lower CT). (B) Deter-
mination of a standard curve: standard fragments were assayed using triplicate mea-
surements for 101, 102, 103, 104, 105, and 106 molecules/assay. The resulting graph (on
a semilogarithmic scale CT:log[standard fragment]) should be a straight line. The R2 value
indicates the reliability of the measurements, which should be above 0.98 if the 101
values are not included. The amplification efficiency (E = 101/m – 1) should be above
80% and t around 40.
90 McAlinden et al.
3.4.2.1. PRIMER SELECTION

Primer selection is best performed using the Primer Express® software sup-
plied by ABI Biosystems. In general, primers should fulfill the following criteria:
1. Length: 18–30 bp.
2. Length of amplified product: 75–150 bp.
3. GC content: 20–80%.
4. Tm: approx 60°C.
5. No poly-T regions (to avoid nonspecific binding to the poly[dT]-tails of cDNAs).
6. No palindromic sequences.
7. Wherever possible, the amplified product should span an exon–intron boundary
(see Note 9).
3.4.2.2. PROBE SELECTION

The TaqMan probes need to be specifically designed using the Primer Express
software, and they can be ordered from a company offering the required modifi-
cations for the assay. They should generally conform to the following criteria:
1. The 5'-end of the probe should be located close to the 3'-end of one of the PCR
primers.
2. Length: 20–30 bp.
3. GC content: 40–60%.
4. Tm of the probe should be 5–10°C above the Tm of the primers to ensure efficient
and complete probe binding.
5. Position of the quencher should be at the 3'-end.
6. No G nucleotide at the 5'-end.
7. Avoid GGGs or CCCs in the probe sequence.
8. No complementarity in between primers and probe.
3.4.2.3. OPTIMIZATION OF ASSAY CONDITIONS

The optimal annealing temperature is usually kept at 60°C according to the
Primer Express program. Generally, two PCR parameters should be tested in
order to identify the optimal amplification conditions, i.e., conditions that cor-
respond to the lowest CT values for the samples tested:
1. Primer concentrations (test 0.05, 0.3, and 0.9 µM for both primers).
2. MgCl2 concentrations (test 3–7 mM in 0.5-mM steps; see Note 10).
3.4.2.4. GENERATION OF STANDARD PROBES (SEE NOTE 6)

In order to calculate absolute copy numbers per assay, one can use gene-
specific standard curves of cloned cDNAs that contain the identical sequence to
be amplified. Alternatively, amplified PCR products can be used: this is less
time-consuming to establish but can be more difficult to manage experimen-
tally. These standard fragments are then assayed in separate tubes to give a
concentration curve, usually 101, 102, 103, 104, 105, and 106 molecules/reaction.
Fig. 5. Quantificative TaqMan analysis for mRNA expression levels of MMP-1

(collagenase A), MMP-8 (collagenase B), and MMP-13 (collagenase C) in normal,
early degenerative, and late-stage osteoarthritic (OA) cartilage. Ratios of each pro-
tease normalized to GAPDH are shown. Stars indicate the significance levels com-
pared to normal articular cartilage (***, p < 0.001).
The resulting CT values should produce a straight line result on a semiloga-

rithmic plot that can subsequently be used to calculate the exact copy number
of cDNA present in a sample (Fig. 4B). Additionally, the parameters of the
standard curve reflect assay reliability (R should be above 0.98), and amplifi-
cation efficiency (m should optimally be between –3.3 and –3.8) and are also
indicative of concentration errors in the standard stocks (t should be about 40)
(Fig. 4B).
3.4.3. Determination of Collagenase mRNA Expression by Real-Time
Quantitative PCR: Experimental Protocol
In order to understand the molecular basis of osteoarthritic cartilage degen-
eration, knowledge of proteases that are able to cleave the main constituent of
adult articular cartilage, collagen type II, are of particular interest. Such candi-
date molecules are collagenases A, B, and C (MMP-1, MMP-8, and MMP-13,
respectively). Expression data obtained by real-time PCR using the protocol
outlined in the following steps clearly support the hypothesis that an important
player in osteoarthritic cartilage degradation is MMP-13 (Fig. 5) (6,7).
92 McAlinden et al.
The sequences of primers and probes, as well as the optimal assay conditions,
for matrix metalloproteinases (MMPs)-1, -8, and –13, as well as a number of
other genes relevant to cartilage function, are shown in Table 2. For standardiza-
tion purposes, ratios relative to GAPDH were calculated. The primers (obtained
from MWG Biotech, Germany) and TaqMan probes (obtained from Eurogentec,
Belgium) were designed using Primer Express software. In order to obtain quan-
titative results, gene-specific standard PCRs using sequence-specific control
probes were performed in parallel (see Subheading 3.4.2.4.). All experiments
were performed in triplicate to minimize inaccurate measurements caused by
pipeting errors (see Note 7). The recommended assay volume is 50 µL, but 25 µL
can also be used to save on expensive material. The experimental protocol is
described in the following sections.
3.4.3.1. REVERSE TRANSCRIPTION

Reverse transcription is performed as described for conventional PCR (see
Subheading 3.1.3.1.). Again, we prefer the use of random oligo primers to
avoid a bias toward 3'-end amplification.
3.4.3.2. TAQMAN ASSAY

1. Preparation of a pre-master mix containing all reagents apart from buffer, Taq
DNA polymerase, and cDNA. This can be stored at –20°C for months in, e.g.,
500-µL aliquots:
a. Upstream primer: 0.25–4.5 µM (optimized according to Subheading 3.4.2.3.).
b. Downstream primer: 0.25–4.5 µM (optimized according to Subheading
3.4.2.3.).
c. TaqMan probe: 0.5 µM.
d. dNTPs: dATP, dTTP, dCTP and dGTP (1 mM of each).
e. ROX buffer: 3 µM.
f. MgCl2: 3–7 mM (use the optimized concentration according to Subheading
3.4.2.3.).
g. Adjust the final volume with H2O.
2. Final master mix for a 25-µL reaction. This is prepared immediately prior to the
experiment (see Note 11):
a. Premaster-mix: 5 µL.
b. 10X polymerase buffer: 2.5 µL.
c. Taq polymerase (5 U/µL): 0.1 µL.
d. Adjust volume to 20 µL with H2O.
3. PCR assay: Usually we use two-step PCR, i.e., the annealing and extension steps
are performed at the same (low) temperature to ensure efficient binding of the
probe to the template during amplification.
a. Add 5 µL of the diluted cDNA to 20 µL master mix. (Dilution of the cDNA

depends on the sensitivity of the assay.)
b. Initial denaturation: 94°C, 4 min (If hot-start enzyme is used, the denaturation
step should be carried out for 10 min; see Note 12.)
c. 40 cycles of: (i.) annealing and extension (Note 13): 60°C, 1 min; and (ii.)
denaturation step: 94°C, 30 s.
3.4.3.3. EVALUATION OF THE REACTION QUALITY

To analyze the quality of the TaqMan reaction, the following parameters
should be considered:
1. Standard deviation of the replicates should be below 0.16. The variations between
replicates are mainly owing to pipeting errors, which can be minimized using mas-
ter mixes.
2. Amplification efficiencies can be calculated: AE = 10–1/m – 1 (Fig. 4B). The
efficiency should be above 70% (optimally above 85%).
3. The mean squared error (MSE) curve displayed in the multicomponent view
should be flat and below 200.
3.5. Normalization of Gene Expression Data

Present technology does not permit the measurement of absolute numbers
of mRNA species within specimens. Therefore, it is necessary to standardize
relative values to internal cellular standards such as the commonly known
housekeeping genes. However, this is dependent on the fact that the expression
of housekeeping genes, e.g., GAPDH or β-actin, does not change significantly
with respect to the biological condition being studied. This assumption, how-
ever, may not be correct, at least in some conditions, although regulation of
these genes appears to be low compared with other genes (Fig. 6); (8). Thus,
for defined cell types such as articular chondrocytes analyzed in similar condi-
tions it is a suitable method. It was suggested recently that multiple housekeep-
ing genes could be used in order to reduce the risk of regulation of
housekeeping genes; however, this is much more time consuming, and signifi-
cant benefits have not been reported to date.
3.6. Conclusion
Quantitative online PCR is one of the most powerful tools to evaluate gene
expression levels. However, care should to be taken with respect to selecting
the normalization procedure. Despite being rather expensive, the high sensitiv-
ity levels of real-time PCR renders this method appealing if one works with
cartilage, from which the isolation of large amounts of RNA is difficult (9).
This is particularly the case if one is interested in analyzing focal lesions such
as those found in osteoarthritic cartilage degeneration (10).
94
Table 2
Nucleotide Sequences of Primers and Probes for Real-Time PCR Quantification
Experiments of Genes Relevant to Cartilage and Chondrocyte Function
Acc. MgCl2
Gene no. Primers Probe conc. (mM) Ref.
β-actin M10277 Up: GCCCTGAGGCACTCTTCCA AGTTTCGTGGATGCCACAGGACTCCAT 6

Low: TTGCGGATGTCCACGTCA
ADAMTS1 NM_006988 Up: GCCAAAGGCATTGGCTACTTC CGTTTTGCAGCCCAAGGTTGTAGATGGT 7 12
Low: GTGGAATCTGGGCTACATGGA
ADAMTS4 AF148213 Up: TGCCCGCTTCATCACTGA ACAGTGCCCATAGCCATTGTCCAGGA 4 6
94
(Aggase1) Low: CAATGGAGCCTCTGGTTTGTC

ADAMTS5 AF142099 Up: CGCTGCCACCACACTCAA AAGTGGCAGCACCAACACAACCAGC 4.5 6
(Aggase2) Low: CGTAGTGCTCCTCATGGTCATCT
Aggrecan NM_013227 Up: ACTTCCGCTGGTCAGATGGA CCATGCAATTTGAGAACTGGCGCC 6 10
Low: TCTCGTGCCAGATCATCACC
Chitinase39 NM_004000 Up: CCTCCTGTCCTTTGACTTCCAT TTGGGAAAAGCCCCTTATCACTG 3.5 13
Low: CCTTGCTCAGAGGGCTGTTG
Chitinase40 NM_001276 Up: CGGAGCCACAGTCCATAGAATC CGGCCAGCAGGTCCCCTATGC 4.5 13
Low: CGTCGTATCCTACCCACTGGTT
Col1A1 NM_000088 Up: AGGGCCAAGACGAAGACATC AATCACCTGCGTACAGAACGGCCTCA 6.5 11
McAlinden et al.
Low: AGATCACGTCATCGCACAACA
Col 2A1 NM_001844 Up: CAACACTGCCAACGTCCAGAT ACCTTCCTACGCCTGCTGTCCACG 5.5 10
Low: CTGCTTCGTCCAGATAGGCAAT
mRNA Expression in Articular Chondrocytes
Acc. MgCl2
Gene no. Primers Probe conc. (mM) Ref.
ColIIA L10347 Up: AGAGGTATAATGATAAGGATGTGTGGAAG CACAGACACAGATCCGGCAGGGC 6 11

Col2-2.exon Low: GTCGTCGCAGAGGACAGTCC
Col 3A1 NM_000090 Up: TGGCTACTTCTCGCTCTGCTT TTCTGGCTTCCAGACATCTCTATCCGCAT 5 11
Low: CGGATCCTGAGTCACAGACACA
Col 10A1 NM_000493 Up: TGCTAGTATCCTTGAACTTGGTTCAT ACGCTGAACGATACCAAACGCCCAC 5.5 11
Low: CTGTGTCTTGGTGTTGGGTAGTG
GAPDH NM_002046 Up: GAAGGTGAAGGTCGGAGTC CAAGCTTCCCGTTCTCAGCC 5.5 9
Low: GAAGATGGTGATGGGATTTC
95
MMP1 NM_002421 Up: CTGTTCAGGGACAGAATGTGCT ACGGATACCCCAAGGACATCTACAGCTCC 6.5 6

Low: TCGATATGCTTCACAGTTCTAGGG
MMP3 NM_002422 Up: TTTTGGCCATCTCTTCCTTCA AACTTCATATGCGGCATCCACGCC 4 6
Low: TGTGGATGCCTCTTGGGTATC
MMP8 NM_002424 Up: GCTCGCTCACTCCTCTGACC TGGTGCCTTGATGTATCCCAACTATGCTTTC 6 14
Low: CATCTTGAGGGAGTGAGTAGTTGCT
MMP13 NM_002427 Up: TCCTCTTCTTGAGCTGGACTCATT TCCTCAGACAAATCATCTTCATCACCACCAC 7 6
Low: CGCTCTGCAAACTGGAGGTC
MMP14 NM_004995 Up: TGCCTGCGTCCATCAACACT AAGACGAATTTGCCATCCTTCCTCTCGT 7 6
Low: CATCAAACACCCAATGCTTGTC
SOX9 Z46629 Up: ACACACAGCTCACTCGACCTTG TTAGGATCATCTCGGCCATCGTCGC 7 3
Low: GGAATTCTGGTTGGTCCTCTCTT
95
96
96
Fig. 6. Demonstration of the variation in expression levels of a panel of housekeeping genes (Human endogenous con-
McAlinden et al.
trol plate, Applied Biosystems) in three stimulation experiments as measured with the high-precision TaqMan device:
human articular chondrocytes were cultured with or without insulin-like growth factor (IGF; 250 ng/ml), interleukin-1β
(Il-1β; 10 ng/ml), or tumor necrosis factor-α (TNF-α; 10 ng/ml). IPC, internal positive control; 18S, 18S rRNA; PO, acidic
ribosomal protein; βA, β-actin; CYC, cyclophilin; GAPDH, GAPDH; PGK, phosphoglycerokinase; β2m, β2-microglobulin;
GUS, β-glucuronidase; HPRT, hypoxanthine ribosyl transferase; TBP, transcription factor II D, TATA binding protein;
TfR, transferrin receptor. Y-axis, difference in number of PCR cycles, corresponding to log 2 ratio of expression.
4. Notes
1. Basically, oligo[dT] primers (starting from the 3'-end of the mRNAs) and ran-
dom oligonucleotide primers (starting in principle from any position of the
mRNA) can be used for reverse transcription. For practical reasons, sequence-
specific primers are generally not used, especially if one is interested in estimat-
ing levels of several gene products in parallel. Generally, reverse transcription is
not always complete, and so using oligo[dT] primers will result in amplification
of products with a bias toward the 3'-end. This implies that the TaqMan-ampli-
fied products should be located near the 3'-end, which might cause problems with
some genes. For this reason, we prefer to use random oligonucleotide primers
during reverse transcription.
2. Different types of reverse transcriptase enzymes such as MMLV and avian myelo-
blastosis virus (AMV) are available commercially from different companies. Over-
all, there is not much difference between these products, although the optimal
reaction temperatures differ between MMLV (37°C) and AMV (up to 58°C). We
usually use an MMLV RT, RNase H minus variant from Promega.
3. The estimation of absolute values of mRNA molecules present in an analyzed
probe (also expressed as a ratio of molecules/GAPDH) is not possible using this
method even if one uses titration curves with standard probes. Realistically, the
values obtained are semiquantitative estimations of the cDNA amounts present in
the analyzed sample. This is mostly owing to different efficiencies of the reverse
transcription reaction for different genes. Another issue is the estimation of abso-
lute numbers of a given expression product per cell. This would require a standard
of known abundance per cell (e.g., housekeeping gene), which is not yet available,
as discussed in Subheading 3.5.
4. It is difficult to estimate protein levels or enzymatic activity based solely on mRNA
measurements of a particular gene. For example, proteins may have long half-
lives, as is the case for cartilage collagens, which represent approx 80% of carti-
lage dry weight. Although abundant in the extracellular matrix, collagen
expression at the mRNA level is hardly detectable in mature chondrocytes (11).
Alternatively, different mRNAs might not be translated with similar efficiencies.
5. In addition to the TaqMan technology, other real-time PCR approaches are avail-
able. Most of these techniques are based on the detection of an intercalating agent
such as SYBR® green, which is detected by its fluorescence and is assumed to be
selectively incorporated into the specific amplification product. These techniques
are not described in detail in this chapter but, in general, they are less reliable as
they do not include the additional specificity control that is provided by the
TaqMan probes, as previously discussed. On the other hand, these techniques do
not require standardized conditions with respect to probe binding (e.g., using
increased Mg concentrations), which means that they can be optimized for signal
specificity in many cases and therefore offer a less expensive and reliable alter-
native. In these assays, it is important to ensure that coamplification of other
transcripts does not occur, as this would result in a nonspecific increase in signal
intensity.
98 McAlinden et al.
6. Theoretically, RNA standards can also be used and added to the RNA sample
before reverse transcription. This allows accurate determination of the number of
RNA molecules present within, e.g., RNA isolated from the cells of interest.
However, one is still not be able to account for different efficiencies of reverse
transcription of different mRNAs. Also, to generate a standard curve one would
need to perform multiple reverse transcriptions per assay with different concen-
trations of standards, which would make the assay time-consuming and expen-
sive under normal conditions. The use of specific polymerases with intrinsic
reverse transcriptase activity such as Tth DNA polymerase is advantageous, but
they are less optimal in the TaqMan assay. Also, the standards in this case should
be different from the probe to be amplified, as they need to be coamplified in one
tube (Multiplex TaqMan PCR, a method not further discussed in this chapter).
7. We usually perform assays in triplicate, which allows one to account for pipeting
errors. If the standard deviation of the results is below 20%, then an average value
can be used for further calculations. Obvious values out of range are excluded
from further calculations. If all three results deviate significantly (more than 50%),
then the measurement is excluded.
8. The ∆∆CT technology theoretically avoids the use of standard curves in order to
quantify relative template amounts. Therefore, this technique is less time-con-
suming and less expensive. In practice, however, this approach is based on sev-
eral assumptions that need to be met rather strictly and that need to be tested
carefully: most importantly, the efficiency of the amplification reaction of both
templates needs to be nearly the same. This is often hard to achieve and even if
this is the case in pilot experiments this is not necessarily true for subsequent
assays according to our experience.
9. The selection of intron spanning primers and probes for TaqMan PCR avoids
amplification of genomic DNA that may be contaminating the specimen to be
measured (either because the primers or the probe are no longer binding, if the
exon–exon boundary is within their sequence, or because the intron is too long to
be amplified during the PCR reaction; therefore the intron should be longer than
2 kb if possible). However, in most cases, genomic DNA is not a major problem
if one is measuring transcripts that are present in reasonable abundance. Also, a
DNAse step during/after RNA isolation might be used to avoid this problem.
10. Since binding of the probe to the template is not stabilized during the amplifica-
tion step (it does not get elongated by the polymerase enzyme), one needs to take
precautions in order to ensure optimal binding. Primarily, the Tm of the probe
should be approx 5–10°C higher than that of the primers. If possible, the high
temperature extension step (i.e., usually carried out at 72°C) should be avoided
by, for example, using a two-step PCR assay combining the annealing and the
elongation step. Furthermore, a higher MgCl2 concentration can be applied in
order to stabilize probe binding.
11. In our experiments we use the ABI PRISM 7700, which provides a 96-well plate
format for the assays using either 50 µL or a 25 µL reaction volume. Recently,
ABI PRISM 7900H was introduced based on a 384-well plate format allowing
the reaction volume to be scaled down to 5 µL. Although this saves on expensive
consumables, it requires a more precise method for pipeting, as offered by spe-
cific robotic systems.
12. Different types of DNA polymerases are available for PCR assays: in principle
one can distinguish conventional DNA polymerases, such as silver star poly-
merase (Eurogentec) and the so-called hot-start DNA polymerases. The latter
enzymes are more expensive but have the advantage that they are only activated
after an extended initial heat treatment (94°C for 10 min). This is of particular
advantage if one intends to store pipetted assays at 4°C for some days before
commencing the experiment. Overall, polymerases should be tested prior to an
experiment in order to select the most sensitive and reliable one. In addition, the
same batch of a chosen polymerase enzyme should be used for all the experi-
ments if possible.
13. The extension time is usually not a critical point in TaqMan analysis as the ampli-
fied products are short (less than 150 bp) and the extension rates of conventional
DNA polymerases exceed 500 bp per min (at 60°C).
Acknowledgments
This work was supported by the BMBF (grant 01GG9824).
References
1. Anonymous (1997) The PCR Technique: Quantitative PCR. Eaton Publishing.
2. Köhler, T., Laßner, D., Rost, A.-K., Thamm, B., Pustowoit, B., and Remke,H.
(1995) Quantitation of mRNA by Polymerase Chain Reaction. Springer-Verlag,
Berlin, Germany.
3.
3 Aigner, T., Gebhard, P. M., Schmid, E., Bau, B., Harley, V., and Pöschl, E.
(2003) Sox 9 expression does not correlate with type II collagen expression in
adult articular chondrocytes. Matrix Biol. 22, 363–372.
4.
4 Gilliland, G., Perrin, S., Blanchard, K., and Bunn, F. H. (1990) Analysis of
cytokine mRNA and DNA: Detection and quantitation by competitive polymerase
chain reaction. Proc. Natl. Acad. Sci. USA 87, 2725–2729.
5.
5 Celi, F. S., Zenilman, M. E., and Shuldiner, A. R. (1993) A rapid and versatile
method to synthesize internal standards for competitive PCR. Nucleic Acids Res.
21, 1047.
6.
6 Bau, B., Gebhard, P. M., Haag, J., Knorr, T., Bartnik, E., and Aigner, T. (2002)
Relative messenger RNA expression profiling of collagenases and aggrecanases
in human articular chondrocytes in vivo and in vitro. Arthritis Rheum. 46, 2648–
2657.
7.
7 Mitchell, P. G., Magna, H. A., Reeves, L. M., et al.. (1996) Cloning, expression,
and type II collagenolytic activity of matrix metalloproteinase-13 from human
osteoarthritic cartilage. J. Clin. Invest. 97, 761–768.
8.
8 Zien, A., Aigner, T., Zimmer, R., and Lengauer, T. (2001) Centralization: a new
paradigm for the normalization of gene expression data. Bioinformatics 17,
S323–S331.
100 McAlinden et al.
9.
9 McKenna, L. A., Gehrsitz, A., Soeder, S., Eger, W., Kirchner, T., and Aigner, T.
(2000) Effective isolation of high quality total RNA from human adult articular
cartilage. Anal. Biochem. 286, 80–85.
10.
10 Gehrsitz, A., McKenna, L. A., Soeder, S., Kirchner, T., and Aigner, T. (2001)
Isolation of RNA from small human articular cartilage specimens allows quantifi-
cation of mRNA expression levels in local articular cartilage defects. J. Orthop.
Res. 19, 478–482.
11.
11 Gebhard, P. M., Gehrsitz, A., Bau, B., Söder, S., Eger, W., and Aigner, T. (2003)
Quantification of expression levels of cellular differentiation markers does not
support a general shift in the cellular phenotype of osteoarthritic chondrocytes.
J. Orthop. Res. 21, 96–101.
12.
12 Wachsmuth, L., Bau, B., Fan, Z., Pecht, A., Gerwin, N., and Aigner, T. (2003)
ADAMTS-1 is a gene product of articular chondrocytes in vivo and in vitro and is
down-regulated by Il-1β. J. Rheumatol. 31, 315–320.
13.
13 Knorr, T., Obermayr, F., Bartnik, E., Zien, A., and Aigner, T. (2003) YKL-39
(chitinase 3-like protein 2), but not YKL-40 (chitinase 3-like protein 1) is
up-regulated in osteoarthritic chondrocytes. Ann. Rheum. Dis. 62, 995–998.
14. Stremme, S., Duerr, S., Bau, B., Schmid, E., and Aigner, T. (2003) MMP-8 is
only a minor gene product of human adult articular chondrocytes of the knee.
Clin. Exp. Rheumatol. 21, 205–209.
RNA Extraction From Cartilage 101
8
RNA Extraction From Cartilage
Frédéric Mallein-Gerin and Jérôme Gouttenoire
Summary
The direct isolation of RNA from cartilage has often proved difficult owing to a number of
factors. Cartilage has a low cell content and contains an extracellular matrix rich in proteoglycans,
which copurify with the RNA as they are large and negatively charged macromolecules. In our
laboratory, we are interested in searching for genes differentially expressed in chondrocytes in
diverse in vivo situations, for instance during maturation of chondrocytes in the growth plate or
during cartilage degeneration. We found that treatment by proteinase K in 1 M guanidinium
isothiocyanate prior to cesium trifluoroacetate ultracentrifugation was crucial to increase the yield
and purity of RNA extracted from cartilage matrix. This protocol indeed led to reproducible pat-
terns of differential display reverse transcriptase-polymerase chain reaction (RT-PCR) and should
be useful for identifying genes differentially expressed by chondrocytes in situ.
Key Words: RNA; extraction; cartilage; chondrocyte; RT-PCR.
1. Introduction
The use of molecular biology techniques has helped elucidate the metabolism
of normal and many pathological processes. Application of these techniques to
human adult articular cartilage has been hampered by a number of factors. Carti-
lage has a low cell content and a highly crosslinked extracellular matrix contain-
ing a high concentration of proteoglycans entrapped in a collagenous network.
Separation of RNA from the aggregating proteoglycans poses a problem because
both classes of molecules are extremely large and highly negatively charged.
Therefore, most gene expression studies are based on RNA extracted from cul-
tured chondrocytes. In situ hybridization can represent an alternative technique
for studying chondrocyte gene expression in cartilage, but this technique is not
quantitative. In our laboratory, we are interested in studying gene expression
during cartilage development or progression of osteoarthritis in human and dif-
ferent animal models. We describe here a reliable protocol, developed by modi-
101
102 Mallein-Gerin and Gouttenoire
fying existing procedures, to isolate DNA-free RNA from cartilage. Purified total
RNA is suitable for analysis of gene expression by quantitative reverse tran-
scriptase-polymerase chain reaction (RT-PCR) or differential display RT-PCR.
2. Materials
To prevent RNase contamination, gloves are worn when handling tissue and
materials. RNase-free glassware and plasticware are used. Solutions are made
with diethyl pyrocarbonate (DEPC)-treated water and autoclaved.
2.1. Equipment for RNA Extraction
1. Mortar and pestle.
2. Liquid nitrogen.
3. Ultracentrifuge (e.g., Beckman).
4. Rotor (e.g., SW60Ti).
2.2. Reagents for RNA Extraction

1. Guanidinium isothiocyanate (GIT).
2. Homogeneization buffer: 4 M GIT, 25 mM sodium citrate, pH 7.0, 0.5% N-lauroyl-
sarcosine, 0.1% β-mercaptoethanol.
3. Phenol (Rnase-free).
4. Chloroform/Isoamyl alcohol (24:1).
5. Proteinase K.
6. Cesium trifluoroacetate (CsTFA).
7. RNase-free DNase (Promega).
8. Rnase inhibitor RNasin (PE Applied Biosystems).
9. Sodium acetate.
10. DEPC-treated water: dissolve 1 g DEPC in 1 L distilled water, and then auto-
clave for 40 min at 128°C. (DEPC inactivates Rnases.)
3. Methods
3.1. RNA Extraction
1. Freeze and store cartilage samples (50–150 mg) into liquid nitrogen.
2. Reduce the frozen samples to powder with a mortar and pestle previously cooled
in liquid nitrogen.
3. Isolate total RNA by using a combination and modification of GIT procedures
(1,2). Homogenize the frozen samples in 5 mL of homogeneization buffer for 3 h
at 4°C.
4. Phenol/chloroform extraction: add to the sample 5 mL of phenol and 2.5 mL of
chloroform/isoamyl alcohol. Mix by vortex and centrifuge at 10,000g for 1 h at
4°C. Carefully remove the upper phase.
5. Precipitation: add 1/10 vol of 3 M sodium acetate, pH 5.2, and 2 vol of frozen
100% ethanol. Mix and incubate at –20°C overnight. Centrifuge at 10,000g for
30 min at 4°C.
RNA Extraction From Cartilage 103
Fig. 1. Agarose-gel resolution patterns of total RNA isolated from bovine cartilage
without (A) or with (B) CsTFA centrifugation. (A) Frozen cartilage samples were
homogeneized in a solution containing 4 M GIT. After phenol-chloroform extraction
and precipitation, the pellets were resuspended in 1 M GIT with 200 µg/mL proteinase
K and incubated at 40°C until complete dissolution. The samples were again phenol-
chloroform extracted and precipitated. RNA preparations were electrophoresed in
formaldehyde-agarose (1%) minigel and stained with ethidium bromide (lane 1) or
transferred to nylon membranes for Northern blot hybridization with a type II collagen
probe (lane 2). Note that this RNA is suitable for a good-quality hybridization. After
ethidium bromide staining, the sample shown in (lane 1) was further stained with tolui-
dine blue. Toluidine blue reveals contamination of proteoglycans (PGs) in an RNA
preparation (lane 3). (B) Electrophoresis and toluidine blue staining of RNA isolated
as indicated in Subheading 3.1. Note that PGs have been eliminated by ultracentrifu-
gation. The positions of the 28S, 18S, and 5S ribosomal RNAs are shown.
6. Resuspend the pellets in 0.8 mL of 1 M GIT with 200 µg/mL proteinase K, and
incubate at 40°C until complete dissolution (see Note 1).
7. Adjust GIT concentration to 4 M by adding 1.2 mL of 6 M GIT, layer the samples
on a cushion of CsTFA with a density of 1.6 g/mL, and ultracentrifuge in a
Beckman SW60Ti rotor at 88,000g for 18 h at 18°C.
8. To eliminate possible traces of genomic DNA, resuspend RNA pellets in a total
volume of 200 µL for digestion with 6 U of RNase-free DNase in the presence of
80 U of RNasin, for 30 min at 37°C.
9. Finally recover total RNA after a further phenol-chloroform extraction and sodium
acetate precipitation. Resuspend the pellet in 10–15 µL DEPC-treated water and
assay 1 µL for concentration and purity of RNA by measuring A260/A280.
4. Notes
1. For our preliminary analyses of gene expression by RT-PCR in human or animal
cartilages, we used classical protocols described in the literature for RNA extrac-
tion. We obtained extremely low yields of RNA, and our RT-PCR amplification
reactions were poorly reproduced, very likely because of contamination by remain-

ing proteoglycans. We found that a treatment by proteinase K in 1 M GIT prior to
CsTFA ultracentrifugation was crucial to increase the yield and purity of RNA
extracted from human, bovine, mouse, rabbit, or chick cartilage. An example is
shown in Fig. 1. In particular, this method led to reproducible patterns of conven-
tional PCR after reverse transcription of minute amounts of RNA, which is espe-
cially convenient for human samples, which are difficult to obtain in high quantities
(3,4). Moreover, this protocol allowed us to obtain reproducible patterns of differ-
ential display RT-PCR and should be useful for identifying genes differentially
expressed by chondrocytes in situ (5).
References
1 Chomczinski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by
1.
acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162,
156–159.
2
2. Adams, M. E., Huang, D. Q., Yao, L. Y., and Sandell, L. J. (1992) Extraction and
isolation of mRNA from adult cartilage. Anal. Biochem. 202, 89–95.
3
3. Bluteau, G., Labourdette, L., Ronzière, M. C., et al. (1999) Type X collagen in
rabbit and human meniscus. Osteoarthritis Cartilage 7, 498–501.
4
4. Bluteau, G., Conrozier, T., Mathieu, P., Vignon, E., Herbage, D., and Mallein-
Gerin, F. (2001) Matrix metalloproteinase-1, -3, -13 and aggrecanase-1 and -2 are
differentially expressed in experimental osteoarthritis. Biochim. Biophys. Acta
1526, 147–158.
5. Bluteau, G., Gouttenoire, J., Conrozier, T., et al. (2002) Differential gene expres-
sion analysis in a rabbit model of osteoathritis induced by anterior cruciate liga-
ment (ACL) section. Biorheology 39, 247–258.
In Situ Hybridization in Cartilage 105
9
Gene Expression Analysis in Cartilage
by In Situ Hybridization
Frédéric Mallein-Gerin and and Jérôme Gouttenoire
Summary
In situ hybridization allows detection and localization of specific nucleic acid sequences
directly within a cell or tissue. We present an in situ hybridization protocol using double-
stranded DNA or single-stranded RNA probes labeled with [32P] to localize and visualize the
temporal and spatial distribution of cartilage-characteristic mRNAs. Probes labeled with this
high-energy isotope provide good resolution at the tissue level with relatively low background;
as a result of the probes that can be obtained that have a higher specificity to emulsion activity,
very short exposure times are required.
Key Words: In situ hybridization; gene; cartilage; chondrocyte; mRNA.
1. Introduction
Rather than analyze an RNA sample after its extraction from a heteroge-
neous cell population, detection of RNA by in situ hybridization identifies
specific cells that contain particular messages and, moreover, makes it pos-
sible to determine biochemical and morphological characteristics of the same
cells. In vitro, it helped us to determine that aggrecan and type II collagen
gene expressions are not necessarily correlated with shape changes during
chondrocyte dedifferentiation (1). In vivo, this technology complements and
supplements gene expression analysis by Northern blotting or reverse tran-
scriptase-polymerase chain reaction (RT-PCR), in the sense that it can dis-
tinguish two types of situation: one in which a gene is expressed in all cells at
a low level and one in which a gene is expressed in a few cells at a high level.
105
2. Materials
2.1. Equipment for In Situ Hybridization
1. 60°C oven.
2. Heating block.
3. Embedding molds.
4. Microscope slides.
5. Microtome.
6. Vacuum dessicator.
7. Slide racks.
8. Water bath.
9. Facilities for handling radioactivity.
10. Darkroom.
2.2. Reagents for In Situ Hybridization

1. 4% Paraformaldehyde (PFA) fixative.
2. Paraffin wax (e.g., Paraplast).
3. Ethanol.
4. Xylene.
5. Triethanolamine.
6. Acetic anhydride.
7. Proteinase K solution: proteinase K in 50 mM Tris-HCl, pH 7.6, 5 mM ethylene-
diamine tetraacetic acid (EDTA). Concentrations of proteinase K are adjusted
from 1 to 10 µg/mL for optimal signal and tissue preservation, depending on the
sample (embryonic or adult cartilage).
8. Hybridization solution: 50% formamide, 10 mM Tris-HCl, pH 7.0, 0.15 M NaCl,
1.0 M EDTA, 1X Denhardt’s solution, 80 µg/mL denatured salmon sperm DNA,
500 µg/mL yeast tRNA, 10% dextran sulfate.
9. Phosphate-buffered saline (PBS).
10. 1X SSC: 0.15 M NaCl, 0.015 M trisodium citrate.
11. Autoradiographic emulsion (e.g., Kodak NTB-3 nuclear track emulsion).
12. Developer (e.g., Kodak D-19 developer).
13. Fixer (e.g., Kodak rapid fixer).
14. [32P]dCTP (3000 Ci/mmol).
3. Methods
3.1. In Situ Hybridization
1. Place dissected tissues in glass vials. Allow fixation of the samples in freshly
prepared PFA. Optimal fixation time is that which gives good morphology as
well as good signal-to-noise ratio after in situ hybridization and must be deter-
mined empirically for each sample.
In Situ Hybridization in Cartilage 107
2. Dehydrate samples in a graded series of ethanol, clear in xylene, embed in

56–57°C Paraplast, section at 7 or 10 µm, and mount the sections on micro-
scope slides.
3. Dry sections overnight on a slide warmer at 37°C, and store at 4°C in the pres-
ence of Drierite until hybridization is performed.
4. Deparaffinize the sections, prior to hybridization, in two changes of xylene for
15 min each. Wash twice in ethanol for 5 min, and allow to air-dry.
5. Treat sections with 4% formaldehyde in PBS for 10 min. Wash in three changes of
PBS for 5 min each and two changes of 70% ethanol for 5 min each, and let dry.
6. Treat the dried tissue sections with proteinase K solution for 10 min at room
temperature, and then rinse in PBS.
7. Treat sections with 4% formaldehyde in PBS for 20 min, wash in two changes of
PBS for 5 min each, treat with 0.1 M triethanolamine for 5 min, and incubate for
10 min in 0.25% acetic anhydride in 0.1 M triethanolamine solution that was
prepared immediately before use.
8. Following acetylation, wash sections twice in 2X SSC for 5 min each, twice in
70% ethanol for 5 min each, and once in 95% ethanol for 5 min, and then air dry
prior to addition of the hybridization solution.
9. Add to the hybridization solution the appropriate labeled probe (see Note 1).
10. Perform hybridization by applying 20- to 40-µL aliquots of the hybridization
solution containing 250,000 cpm of probe onto the sections and covering them
with siliconized cover slips, the edges of which are then sealed with rubber ce-
ment. Incubate the slides at 50–52°C for about 20 h.
11. After hybridization, remove the cover slips in 2X SSC, and consecutively wash
the sections twice in 2X SSC for 10 min each at room temperature, once in 0.5X
SSC for 10 min at 50°C, and three times in 0.1X SSC for 15 min each at 50°C.
Then dehydrate the sections in two changes of 70% ethanol for 5 min each and
one change of 95% ethanol for 5 min and let dry.
12. For autoradiography, dip the slides in autoradiographic emulsion diluted 1:1 with
water at 42°C. After drying vertically at room temperature for at least 1 h, expose
the slides in the dark at 4°C in the presence of Drierite for 3–8 d.
13. Develop autoradiographs for 3 min at 18°C, fix, and wash in running tap water.
Stain the slides with hematoxylin, dehydrate them with a graded series of
alcohols, clear in xylene, and mount with Permount.
4. Notes
1. For our in situ hybridization experiments, we obtained very similar results using
either the double-stranded DNA probes or single-stranded RNA probes (2), each
of which offers certain advantages. For instance, benefits of using single-stranded
RNA probes are that posthybridization RNase digestion can be used to eliminate
mismatched duplexes resulting from the hybridization of the probe to homolo-
gous regions of other mRNAs. On the other hand, double-stranded DNA probes
are easier to prepare and utilize than RNA probes, and lower hybridization and
washing conditions are required to provide sufficiently stringent conditions to
limit crosshybridization. An example is shown in Fig. 1.
Fig. 1. Autoradiographs of frontal sections through an hindlimb of a 17.5-d-old

mouse embryo that were hybridized with 32P-labeled probes. At this stage, cartilage
rudiments form prospective long bones. (A) Brightfield and corresponding darkfield
(B) images of a section hybridized with a type X collagen DNA probe. (C) Darkfield
image of a section adjacent to that in A and B that was hybridized with a type II
collagen DNA probe. Note the intense accumulation of silver grains representing
hybridizable type II collagen mRNA sequences over the well-differentiated cartilage
rudiments and the restricted hybridization of type X collagen mRNA sequences in the
regions where chondrocytes undergo hypertrophic maturation.
References
1.
1 Mallein-Gerin, F., Ruggiero, F., and Garrone, R. (1990) Proteoglycan core pro-
tein and type II collagen gene expressions are not correlated with cell shape
changes during low density chondrocyte cultures. Differentiation 43, 204–211.
2. Mallein-Gerin, F., Kosher, R. A., Upholt, W. B., and Tanzer, M. L. (1988) Tem-
poral and spatial analysis of cartilage proteoglycan core protein gene expression
during limb development by in situ hybridization. Dev. Biol. 126, 337–345.
Microarray Analysis of Cartilage and Chondrocytes 109
10
Analysis of Differential Gene Expression
in Healthy and Osteoarthritic Cartilage
and Isolated Chondrocytes by Microarray Analysis
Thomas Aigner, Joachim Saas, Alexander Zien,

Ralf Zimmer, Pia M. Gebhard, and Thomas Knorr
Summary
The regulation of chondrocytes in osteoarthritic cartilage and the expression of specific
gene products by these cells during early-onset and late-stage osteoarthritis are not well charac-
terized. With the introduction of cDNA array technology, the measurement of thousands of
different genes in one small tissue sample can be carried out. Interpretation of gene expression
analyses in articular cartilage is aided by the fact that this tissue contains only one cell type in
both normal and diseased conditions. However, care has to be taken not to over- and misinter-
pret results, and some major challenges must be overcome in order to utilize the potential of
this technology properly in the field of osteoarthritis.
Key Words: Microarray; differential gene expression; cartilage; chondrocyte; osteoarthritis.
1. Introduction
The regulation of cartilage cells (chondrocytes) in osteoarthritic cartilage and
the expression of specific gene products by these cells during early-onset and
late-stage osteoarthritis are not well characterized. However, with the introduc-
tion of cDNA-array technology (1), the measurement of thousands of different
genes in one small tissue sample can be carried out. Interpretation of gene
expression analyses in articular cartilage is aided by the fact that this tissue
contains only one cell type in normal and diseased conditions. Thus, changes in
gene expression in osteoarthritic cartilage will be an effect produced only by the
chondrocytes and not by other inflammatory cells, which may be present in the
surrounding synovial fluid, for example.
109
110 Aigner et al.
2. Materials
1. 12X MES (2-[N-Morpholino]ethanesulfonic acid), 1.22 M, pH 6.6, 0.89 M [Na+]
(Sigma, St. Louis, MO).
2. 5X RNA fragmentation buffer: 200 mM Tris-acetate, pH 8.1, 500 mM KOAc,
150 mM MgOAc.
3. Phase lock gel (Brinkman Instruments, Westbury, NY).
4. DNAse (RNAse free; Qiagen, Hilden, Germany).
5. 20X SSPE: 3 M NaCl, 0.2 M NaH2PO4, 0.02 M EDTA.
6. Diethyl pyrocarbonate (DEPC; Carl Roth, Karlsruhe, Germany).
7. T7 (dT)24 primer (5'-GGCCAGTGAATTGTAATACGACTCACTATAGGGAG
GCGG(dT)24-3' and SuperScript Choice System for cDNA synthesis (Invitrogen
Life Technologies, Karlsruhe, Germany).
8. Herring sperm DNA (Promega, Madison, WI).
9. RNeasy kit (Qiagen).
10. Human cancer 1.2 cDNA arrays (BD Bioscience Clontech, Heidelberg, Germany).
11. Salmon testes DNA (Sigma-Aldrich Chemie GmbH, München, Germany).
12. Wash solution 1: 2X SSC (15 mM NaCl/1.5 mM Na3 citrate) ,1% sodium dodecyl
sulfate (SDS).
13. Wash solution 2: 0.1X SSC (0.75 mM NaCl / 0.075 mM Na3 citrate), 0.5% SDS.
14. [α-32P]dATP: 3000 Ci/mmol, 10 µCi/µL (Amersham, Uppsala, Sweden).
15. Enzo BioArray HighYield RNA transcript labeling kit (Enzo Diagnostics,
Farmingdale, NY).
16. GeneChip® Eukaryotic Hybridization control kit (Affymetrix, Santa Clara, CA).
17. Affymetrix GeneChips® (Affymetrix).
18. Antibody solution mix: 1X MES stain buffer (100 mM MES, 1 M [Na+], 0.05%
Tween-20), 2 mg/mL acetylated bovine serum albumin (BSA; Invitrogen Life
Technologies), 0.1 mg/mL normal goat IgG (Sigma-Aldrich), 3 µg/mL
antistreptavidin antibody (Vector, Burlingame, CA).
19. SAPE (streptavidin/phycoerythrin) solution mix: 1X MES stain buffer (100 mM
MES, 1 M [Na+], 0.05% Tween-20), 2 mg/mL acetylated BSA, 10 µg/mL R-phy-
coerythrin streptavidin; Molecular Probes, Leiden, Netherlands).
20. GeneChip® Fluidics Station 400, GeneChip® Hybridization Oven, GeneChip®
Scanner (Affymetrix).
21. 6800 Freezer/Mill (SPEX CertiPrep, Metuchen, NJ).
22. PhosphorImager (e.g., Molecular Dynamics, Sunnyvale, CA, or Bio-Rad, Her-
cules, CA).
3. Methods
3.1. RNA Isolation
The effective isolation of sufficient amounts of high-quality RNA is of pri-
mary importance for gene expression analyses (e.g., Clontech: 2–50 µg total
RNA; Affymetrix: 5–10 µg total RNA). There are preamplification methods
available that reduce the starting amount of RNA required, but these methods
can cause some technical problems and thus are not yet well established.
3.1.1. Isolation of RNA From Cultured Chondrocytes
The isolation of RNA from chondrocyte cultures can be done using conven-
tional isolation protocols (e.g., Qiagen RNeasy isolation kits).
3.1.2. Isolation of RNA From Normal and Osteoarthritic Adult Cartilage
The isolation of large quantities of good-quality RNA directly from human
adult articular cartilage is difficult owing to two main factors: (1) low cell
number (2–3% chondrocytes per total mass of cartilage tissue) and (2) the pres-
ence of a proteoglycan-rich, highly crosslinked extracellular matrix. The major
proteoglycan in articular cartilage, aggrecan, tends to copurify with the RNA
because it is also a large, negatively charged macromolecule. The following
protocol describes a reliable method for the isolation of reasonably high
amounts of pure RNA from adult articular cartilage (2).
3.1.2.1. TISSUE PREPARATION
Remove cartilage from the joint and freeze immediately in liquid nitrogen
(slow freezing leads to degradation of RNA). Store the cartilage at –80°C until
further use.
3.1.2.2. TISSUE HOMOGENIZATION
This is done in our laboratory using a SPEX CertiPrep freezer mill 6800 (see
Note 1).
1. Precool a vial and tissue grinder apparatus in liquid nitrogen.
2. Fill SPEX CertiPrep 6751 vial with frozen cartilage tissue (approx 2.5 g maximum).
3. Insert two to four vials into the freezer mill. Following a precooling phase (2
min), grind the sample for five cycles of impact (2 min, at a frequency of 10 Hz)
and cooling (2 min).
4. Homogenized tissue can be stored at –80°C prior to RNA isolation.
3.1.2.3. RNA ISOLATION

We use the RNeasy-kit from Qiagen with some modifications to the
manufacturer’s protocol.
1. Mix 1 g of milled cartilage powder with 5 mL of RLT lysis buffer (containing
10 µL/mL β-mercaptoethanol) at room temperature. Vortex to homogeneity
(15 s) and centrifuge at 10,000g for 1 h.
2. To the cleared lysate, add an equal volume of 70% ethanol and vortex for 15 s.
3. Apply aliquots (3 mL) of the cleared lysate/ethanol mix sequentially to Qiagen
RNeasy midi-columns. Centrifuge at 3360g for 1 min and repeat until all the
lysate has been applied to the column (see Note 2).
112 Aigner et al.
4. DNAse digestion: add 3 mL (instead of the recommended 4 mL) of wash buffer

RW1 to the column followed by 80 Kunitz units DNAse I in 210 µL RNAse-free
DNAse buffer. Following digestion for 25 min at room temperature, apply an-
other 3 mL of RW1 washing-buffer to the column and centrifuge at 3360g for 5
min.
5. Wash the column with 2.5 mL of RPE wash buffer and centrifuge twice at 3360g
for 2 min. Centrifuge for a further 5 min to ensure complete removal of ethanol
(present in wash buffer RPE) from the column.
6. Elute the RNA twice in 160 µL aliquots of DEPC-treated water by centrifugation
at 3360g for 3 min.
3.1.2.4. RNA ANALYSIS AND QUANTIFICATION

1. Agarose gel electrophoresis: analyze two aliquots of isolated RNA (1 and 5 µL)
on either 1.2% formaldehyde-agarose denaturing gels or on 1.2% agarose gels.
(Conventional DNA gels are usually sufficient.) Gel electrophoresis will deter-
mine whether the RNA is degraded or whether there is contamination of genomic
DNA as well as provide a rough estimation of RNA concentration.
2. Spectrophotometry: measure an aliquot of RNA at 260 and 280 nm (or, if needed,
perform a full spectrum analysis from 230 to 400 nm). This allows quantification
of RNA concentration as well as RNA purity (i.e., a low OD260/OD280 ratio sug-
gests protein contamination; see also Note 3).
3.2. cDNA Array Analysis

A DNA microarray, or gene-chip, is defined as a matrix of hundreds or thou-
sands of DNA fragments (>200 bp) or oligonucleotides (20–80 bp) bound to a
solid support (e.g., nylon membranes or glass slides). Labeled cDNA from the
tissue or cell type of interest is hybridized to its complementary sequence on the
gene-chip and then the relative abundance of mRNA in the sample is estimated.
The final readout from this hybridization experiment gives what is referred to as
an expression profile of genes in a given sample. A flow chart summarizing the
gene profiling approach is shown in Fig. 1.
3.2.1. Gene Expression Analysis Parameters
Some basic parameters have to be considered when performing gene
expression analysis:
1. Complexity of the array: the number and selection of genes on the array chip will
determine the success in identifiying genes of interest. Gene-chips are available
containing less than 100 different genes (often custom-made according to the
needs of the user) up to more than 10,000 genes.
2. Type of cDNA labeling: radioactive as well as fluorescent (e.g., rhodamine, fluo-
rescein, Cy3, or Cy5) labeling can be used depending on the array type. Utilizing
two different-colored fluorochromes allows direct visualization of samples show-
ing different expression profiles. Radioactive labeling of cDNA offers greater
Fig. 1. Basic workflow scheme of a gene profiling approach. After isolation of total
RNA from the cells or tissues of interest, cDNA is generated by reverse transcription.
Usually, this step is directly combined with labeling of the cDNA. In some protocols
(e.g., Affymetrix) an amplification step is introduced using an additional in vitro tran-
scription step of the cDNA for probe labeling. The labeled cDNA/cRNA is hybridized
to the immobilized DNA on a microarray. After stringent washing the amount of bound
target DNA is determined by counting the radioactivity on the microarray.
sensitivity but lacks the ability to compare two different sets of mRNA pools
directly.
3. RNA quality: the quality of RNA used to make probes is the most important
factor influencing the sensitivity and reproducibility of the hybridization results
(for RNA isolation, see Subheading 3.1.). Contaminated or partially degraded
RNA may lead to high background and thus, inaccurate hybridization patterns.
Both total RNA or poly(A)+RNA have been used successfully as a starting
material for the generation of probes. In our laboratory, we usually use total RNA.
4. cDNA synthesis: the reverse transcription reaction usually starts from the poly(A)-
tail of the oligo(dT)-primed mRNA molecule. Since all oligo(dT)-primed-mRNAs
are not reverse-transcribed with the same efficiency (a process that begins at the
3’end of the RNA molecule), random primers are sometimes preferred since
reverse transcription can be initiated throughout the mRNA molecule. However,
use of random primers can also lead to reverse transcription (and hence labeling)
of irrelevant mRNA species such as ribosomal RNA and transfer RNA; these are
not labeled if oligo(dT)-primers are used. The use of gene-specific primers for
reverse transcription (e.g., Clontech Cancer array) ensures only the synthesis
of cDNAs corresponding to genes present on the particular array. Therefore, the
114 Aigner et al.
hybridization probes created here are significantly less complex than those gener-
ated using oligo(dT) or random primers. This results in increased sensitivity with
a concomitant reduction of nonspecific background in the array experiment.
It should be noted that, since the reverse transcription reaction is not equally
efficient for all genes under analysis, the intensities of hybridized probes (“spots”)
can only be compared for the same gene between different assays and not between
different genes in one assay.
5. Hybridization and posthybridization conditions: these important parameters are
provided by the manufacturer. Particularly, the sensitivity and specificity of
probes are of major concern in all array systems (see Note 4).
3.2.2. Nylon Membrane-Based cDNA Arrays: Experimental Protocol

Nylon membranes may contain hundreds to more than a thousand different
cDNAs. For our studies, we used the Human Cancer 1.2 cDNA AtlasTM arrays
(Clontech), which are nylon membranes containing 1185 cDNAs (a typical
result is shown in Fig. 2A).
3.2.2.1. PROBE SYNTHESIS FROM TOTAL RNA: REVERSE TRANSCRIPTION
The reverse transcription reaction described below converts 2–5 µg of total
RNA into radioactive-labeled first-strand cDNA (one can use 32P-, 33P- or 35S-
labeled deoxyribonucleotides). All components are contained in the Human
Cancer 1.2 cDNA Atlas kit.
Fig. 2. (opposite page) (A) RNA (5 µg total) was isolated from human articular
cartilage and hybridized with the Atlas human cancer 1.2 cDNA array (Clontech). The
detected dots correspond to the amount of labeled probes hybridized to cDNAs repre-
senting the arrayed genes. (B–D) The study of chondrocyte gene expression is always
linked to the main function of this cell type, which is the preservation and turnover of
the cartilage matrix. Thus, matrix components and matrix-degrading proteases were a
focus of interest in this study. In our analysis of the extracellular matrix proteins (21),
we could confirm an absence of cartilage collagen expression in normal cartilage.
mRNA levels of several collagen genes were identified in advanced osteoarthritis,
however (B). With regard to the cartilage matrix-degrading metalloproteinases, MMP-
3 (stromelysin) was surprisingly downregulated in the diseased tissue (C). Instead,
other degradation pathways appeared to be more important involving proteases such
as MMP-2 (gelatinase A) and MMP-13 (collagenase 3) (D). Both MMP-2 and MMP-
13 are known to be involved in terminal breakdown of cartilage collagen fibers. Bars,
means with indicated standard deviations; significance levels: *, p < 0.05; **, p < 0.01;
***, p < 0.001). OA, osteoarthritis.
115
116 Aigner et al.
1. Preparation of the “reaction mix” (carried out at room temperature):

5X reaction buffer (for the specific enzyme) 2.0 µL
10X dNTP mix 1.0 µL
(for dATP labeling, the dNTP mix contains
5 mM each of dCTP, dGTP and dTTP)
[α-32P]dATP (3000 Ci/mmol, 10 µCi/µL) 3.5 µL
Dithiothreitol (DTT; 100 mM) 0.5 µL
Total volume/assay 7.0 µL
2. Preparation of the RNA/primer mix (in a 0.5-mL PCR tube):
5X reaction buffer (for the specific enzyme) 2.0 µL
RNA 2–5 µg (1–2 µL)
Primer mix 1 µL
H2O X µL
Total volume/assay 3 µL
3. Reverse Transcription.
a. Mix the RNA/primer mix by pipeting followed by a brief centrifugation step.
b. Incubate at 70°C for 2 min. (We use a PCR cycler, which allows rapid
cooldown to 50°C.)
c. Incubate at 50°C for 2 min. During this incubation step, add 1 µL murine
Moloney leukemia virus (MMLV) reverse transcriptase and mix by pipeting.
The reaction mix is at room temperature at this stage.
d. Immediately add 7 µL of the reaction mix to 3 µL RNA/primer mix.
e. Incubate at 50°C for 25 min.
f. Stop the reaction by addition of 1 µL 10X termination mix (provided by the
manufacturer).
g. Labeled probes can be stored on ice at 4°C for a few hours, if necessary.
3.2.2.2. HYBRIDIZATION PROCEDURE

The hybridization procedure is performed as described in the manufacturer’s
hybridization protocol (all components are contained in the Human Cancer 1.2
cDNA Atlas kit.):
3.2.2.2.1. Preparation of Hybridization Solution
1. Prewarm 5 mL of BD ExpressHyb solution to 68°C.
2. Heat 0.5 mg of the sheared salmon testes DNA at 95–100°C for 5 min, and then
chill quickly on ice.
3. Mix heat-denatured sheared salmon testes DNA with prewarmed BD ExpressHyb
solution.
4. Store at 68°C until use.
3.2.2.2.2. Preincubation
1. Place the membrane into a bottle filled with deionized H2O. After discarding the
water, the membrane should adhere around the inside of the bottle without creat-
ing air bubbles.
2. Add 5 mL of the hybridization solution prepared as described in Subheading
3.2.2.2. (Ensure that the solution is evenly distributed over the membrane.)
3. Prehybridize the nylon membrane for 30 min with continuous agitation at 68°C.
3.2.2.2.3. Probe Preparation

1. Heat the radiolabeled probe (95–100°C) in a water bath for 2 min.
2. Transfer immediately onto ice for 2 min.
3. Add the labeled probe to the prehybridization solution coating the nylon mem-
brane inside the bottle. Ensure that the probe is equally distributed over the mem-
brane.
4. Hybridize the membrane overnight with continuous agitation at 68°C. If neces-
sary, add an extra 2–3 mL of prewarmed BD ExpressHyb solution in order to
ensure even coverage of the filter during the incubation period.
3.2.2.2.4. Washing Procedure

1. Prewarm wash solution 1 (2X SSC, 1% SDS) and wash solution 2 (0.1X SSC,
0.5% SDS) at 68°C.
2. Carefully remove the hybridization solution.
3. Wash the membrane four times in 200 mL of prewarmed wash solution 1 for 30
min at 68°C with continuous agitation.
4. Wash twice in 200 mL of prewarmed wash solution 2 for 30 min at 68°C with
continuous agitation.
5. Finally, wash twice with 200 mL SSC for 5 min at room temperature.
3.2.2.2.5. Signal Detection

1. Remove the membrane from the bottle and wrap in plastic cling-film to keep it
moist.
2. Expose the membrane array to a storage phosphor screen.
3.2.2.3. IDENTIFICATION OF LABELED, HYBRIDIZED “SPOTS”

Different software tools can be used for the detection and quantification of
hybridized cDNA spots, depending on the array systems. For the Atlas Nylon
cDNA Expression Arrays, we use the Clontech Atlas Image 1.01 software.
Updated versions of this software are now available. It is also useful to analyze
the spots by eye, if possible, in order to confirm the accuracy of the detection
methodology. A spot is considered positive if it is easily located, “roundish,”
and has an intensity more than twofold above the background level.
Other parameters of data acquisition such as technical reproducibility, assay
sensitivity, and signal quantification are not discussed in this chapter. The
reader should see recently published reviews describing these issues in more
118 Aigner et al.
detail (3–6). Importantly, data measurement should be validated by internal

positive and negative control genes. In addition, the results should be repro-
ducible and also confirmed by an independent assay system. These controls are
necessary as data from some microarray assays may be misleading owing to
effects of crosshybridization and low sensitivity, to name a few.
3.2.3. High-Density Oligonucleotide Arrays
(Affymetrix GeneChip® Technology)
Another commonly used microarray platform is the Affymetrix GeneChips®
technology (7,8), which contain up to 1 million distinct oligonucleotide cells
(probe cells) on an area of 1.6 cm2. A probe cell contains millions of copies of an
oligonucleotide (probe), and these are synthesized by photolithography and com-
binatorial chemistry on a glass substrate. Each transcript is measured by a set of
probes (typically between 11 and 20). The probes, 25-mers selected for optimal
hybridization, are subsequences of the target-sequence, which is preferentially
located 600 nucleotides upstream from the polyadenylation site of each gene. In
addition, 25-mer oligonucleotides are also synthesized that contain a mismatch
base at the 13th position for measuring nonspecific crosshybridization. Perfect
match (PM) and mismatch (MM) oligomers constitute a probe pair (Fig. 3).
3.2.3.1. TARGET PREPARATION
Preparation of labeled nucleic acids (known as the target) for hybridization
is described in detail in the Affymetrix Technical Manual. We routinely use
5–10 µg of total RNA (see Subheading 3.1.) and amplify it according to the
method of Eberwine et al. (9).
1. Basically, RNA is reverse transcribed using a T7(dT)24 primer (containing the
phage T7-RNA polymerase promotor at the 5'-end) and 200 U SuperScript II RT
(Invitrogen Life Technologies) at 42°C.
2. Second-strand synthesis is carried out using E. coli DNA polymerase I,
E. coli DNA ligase, and E. coli RNAse H at 16 °C.
3. Blunt-end cDNA fragments are then made using T4 DNA polymerase.
4. The cDNA is purified by phenol-chloroform extraction and centrifuged using
phase lock gels, which facilitates good separation of aqueous, interface and or-
ganic layers.
5. The cDNA is precipitated with ethanol and resuspended in RNAse-free water.
6. The purified cDNA is transcribed in vitro, whereas biotin-labeled UTP and CTP
are incorporated by T7-RNA polymerase (Bioarray HighYield RNA transcript
labeling kit, Enzo Diagnostics).
7. One-half of the amplified RNA is purified using RNeasy spin columns (Qiagen)
and quantitated by spectrophotometry at 260 nm.
8. Twenty micrograms of cRNA are subsequently fragmented by metal-induced
hydrolysis for 35 min at 94°C into fragments with a length of 35–200 nucleotides
using the fragmentation buffer.
Fig. 3. Each gene is represented by multiple probe pairs on a GeneChip. An exem-

plary probe set consisting of 11 probe pairs (perfect match oligos and mismatch oligos)
is shown in the enlarged picture. The fluorescence intensity image of the probe set is
shown at the bottom. The sequences of the perfect match oligo and mismatch oligo for
one probe pair, and the corresponding probe cells in the fluorescence intensity image,
are boxed.
9. The fragmented cRNA is supplemented with biotinylated control oligonucleotide

B2 (50 pM) and with quantitative control standards: biotinylated transcripts bioB,
bioC, bioD, cre (1.5, 5, 25, and 100 pM, respectively).
10. The oligonucleotide B2 provides alignment signals for the placement of a grid
during data analysis of the scanned image. Both the B2-oligo and the control
standards are provided in the GeneChip Eukaryotic hybridization control kit.
11. Fragmented cRNA containing biotinylated control and standards is then adjusted
to a volume of 300 µL by addition of 150 µL 2X hybridization buffer (200 mM
MES, 2 M [Na+], 40 mM EDTA, 0.02% Tween-20), and RNAse-free water.
3.2.3.2. TARGET HYBRIDIZATION
1. Heat the hybridization cocktail at 99°C for 3 min to denature the cRNAs.
2. Equilibrate to 45°C for 5 min, centrifuge (12,000g, 5 min) and fill the cleared
supernatant into the GeneChip cartridge.
3. Hybridize at 45°C for 16 h with mixing on a rotisserie at 60 rpm.
3.2.3.3. WASHING, STAINING, AND SCANNING OF GENECHIPS
1. Remove the hybridization cocktail from the cartridge.
2. Fill the cartridge with nonstringent wash buffer (6X SSPE, 0.01% Tween-20).
The following washing and staining steps are performed using the Affymetrix-
Fluidics Station:
120 Aigner et al.
3. Wash the GeneChip with nonstringent wash buffer at 25°C (10 cycles of 2 mixes/
cycle).
4. Wash with stringent wash buffer (100 mM MES, 0.1 M [Na+], 0.01% Tween-20)
at 50°C (four cycles of 15 mixes/cycle).
5. Stain with SAPE solution for 30 min at 25°C .
6. Wash with nonstringent wash buffer at 25°C (10 cycles of 4 mixes/cycle).
7. Incubate with antibody solution for 10 min at 25°C .
8. Stain with SAPE solution for 10 min at 25°C.
9. Wash the GeneChip with nonstringent wash buffer at 30°C (15 cycles of 4 mixes/
cycle).
10. The gene-chip arrays are then analyzed twice with a confocal scanner (Hewlett-
Packard Gene Array scanner) using an argon laser (λexcitation = 488 nm, λemission =
570 nm).
3.2.3.4. PRIMARY DATA ANALYSIS
1. The array images are acquired and quantitated with Affymetrix Microarray Suite
4.0.1.
2. The images (saved as .dat files) are initially controlled visually, and subsequently
the files for cell intensity (.cel) and chip intensity (.chp) are generated.
3. Next, the scans to be compared are normalized (“globalized”: see Subheading
3.3.). For each gene, the average difference intensity (ADI) is calculated, which
reflects the transcript level of a gene. Essentially, for each probe pair, the mis-
match (MM) intensity is subtracted from the perfect match (PM) intensity, and
the average for all the probe pairs for a specific gene is calculated excluding
outliers (Fig. 3).
4. In addition, a decision matrix is employed to render a gene either absent (A),
present (P), or marginal (M). Data files containing intensity values as well as A
and P information can be exported and processed further using available soft-
ware packages (see also Note 5).
5. In a recent experiment, we analyzed the effect of interleukin-1β (Il-1β) on cul-
tured human articular chondrocytes (Fig. 4B) and found a significant alteration
in gene expression. In comparison, Fig. 4A shows the technical scatter of genes
using identical control RNAs in two independent analyses.
3.3. Normalization
A prerequisite for biologically meaningful comparisons between microarrays
is that the data have to be adjusted to the same scale (i.e., “normalized”). In large-
scale gene expression biotechnology, “globalization” is the most commonly used
normalization method: for each array, all measured values are divided by their
sum (or average). This method is based on the assumption that the amount of
mRNA per cell is constant. However, the sum of all expression signals is often
dominated by the strongest signals (10,11). Such highly expressed genes are most
likely to be regulated as they represent the major expression products of special-
ized cells (e.g., immunoglobulin chains for plasma cells, hemoglobin for erythro-
Fig. 4. (A) The same target material (prepared from T259 RNA of in vitro cultured
primary chondrocytes) was hybridized consecutively onto two HG-U95Av2
GeneChips. The intensities (ADI; see Subheading 3.2.3., Primary data analysis) for
each gene (represented as squares) are plotted for the two scanned GeneChips. High
technical reproducibility is recognizeable since most genes reside on a diagonale. (B)
Diagram showing the effect of IL-1β treatment on primary human chondrocytes in
vitro. RNA from nontreated primary chondrocytes (T259_control) and cells treated
with IL-1β for 24 h (T259_IL1) was labeled and hybridized onto HG-U95Av2
GeneChips.
blasts). For this reason, we recently proposed a method (“centralization”: http://

www.scai.fhg.de/bio/centralization.html) that rests on the weaker assumption
that the regulation of gene expression is well behaved, i.e., that most genes are
either not regulated or only moderately regulated. The choice of normalization
method is important in order to avoid false findings (11).
3.4. Evaluation of Gene Expression Data:
How to Proceed From Data to Understanding
Given reliable data on gene expression levels, the real challenge of func-
tional genomics then has to be addressed, i.e., how to interpret results from a
mass of data and relate them to an understanding of the processes in a biologi-
cal system or disease. Fundamentals of such efforts, as outlined below, include:
1. Pre-existing knowledge of relevant biological systems (genes of known relevance;
an example for cartilage research is outlined in Fig. 2B–D).
2. Ranking genes according to biostatistical significance (p-value-approach).
3. The comparison of gene expression patterns with those found in vivo (evo-devo-
approach) or in vitro models.
122 Aigner et al.
Fig. 5. “Functional genomics” in many cases starts from gene expression analysis,
which yields a huge amount of unstructured primary data. Biostatistical evaluation then
allows one to highlight directly the significance levels of differentially expressed genes
of defined interest. Genes also become a focus of interest if they show significant dif-
ferential regulation or show close clustering to other well-defined genes. Improved com-
putational tools will further allow us to search specifically for biologically relevant
molecular networks, or parts of them. These “new” genes will have to be validated for
their relevance in the respective disease process, e.g., osteoarthritis. Analyzing both
gene expression and function during evolution and development (evo-devo-approach)
as well as pursuing functional assays of the genes in a test tube (molecular approaches),
in vitro (cell culture) or in vivo (animal models, transgenic approaches) will establish
the roles of these molecules in physiology and pathology.
4. Bioinformatical tools (clustering and neighborhood-analyses).

5. A combination of preexisting knowledge and computational processing of obtained
data (i.e., bioinformatical pathway modeling. A possible workflow diagram is
shown in Fig. 5).
6. Final confirmation of a revealed pattern then requires functional validation so
that the molecular networks can be identified.
3.4.1. Search for Novel Genes of Interest

1. The p-value approach. A basic but useful method of analysis is the search for
differentially expressed genes between groups of samples (e.g., normal vs dis-
eased tissue). Parametric tests are most commonly used but are based on the
assumption that the data follow a specific distribution, e.g., the t-test assumes a
normal distribution. Although gene expression levels seem to follow other distri-
butions, transformations have been proposed in order to achieve normal distribu-
tions. Another solution is offered by distribution-free tests, e.g., the rank sum test
and new tests developed specifically for expression data (12,13). Either way, one
should be aware that, owing to the high numbers of genes assayed, low p values
may still lead to many false positives. Therefore derived measures, such as the
false discovery rate (14), may be more helpful in practice.
2. Clustering and neighborhood analyses. In addition to analyses at the single-
gene level, many algorithms are available to cluster genes or tissue samples, or
to perform neighborhood analyses with the whole dataset in an attempt to iden-
tify coregulated genes. In fact, depending on the stringency of the clustering
criteria applied, one can find various numbers of gene clusters such as those
containing most of the ribosomal genes and others containing regulated col-
lagen genes, etc. (15). However, clustering and neighborhood analyses as such
are only of limited value as long as they lack integration with overall knowledge
on the genes involved.
3. Validation of novel genes. Regardless of what successful bioinformatical analysis
is performed, validation of the cellular role and molecular function of a detected
gene is of primary importance. This area of “functional genomics” represents the
most challenging task in terms of study design and manpower required. Even when
the functions of well-characterized gene products are known in other contexts, it
is not guaranteed that they will fulfill similar functions in the system under study.
a. The evo-devo-approach. One way to attempt to understand regulatory pat-
terns of mature cells and tissues in the adult is to exploit the fact that in many
conditions, processes that occurred during evolution and individual develop-
ment are recapitulated. Although an evolutionary model for cartilage has not
really been established, the fetal growth plate of cartilage has served as a
long-standing model in the study of chondrocyte behavior during develop-
ment. In addition, the established chondrocyte phenotypes are mainly based
on observations of events that occur in the growth plate. Furthermore, pro-
cesses known to be central in osteoarthritic cartilage degeneration, such as
matrix anabolism and catabolism and, in particular, cellular differentiation,
are observed within the fetal growth plate. It remains a challenging task to
delineate which patterns are found in which spatiotemporal sequence in the
diseased tissue as well.
b. In vivo and in vitro models. Clearly, in vivo (animal) models are of paramount
importance for any disease area in order to understand the disease processes
clearly and what happens if these processes are modified. Unfortunately, no
generally accepted animal model for osteoarthritis is available at the moment.
Isolated articular chondrocytes (in vitro models) have been known for a long
time to be susceptible to drastic alterations of their gene expression pattern.
This is in response to modulating factors such as cytokines and growth factors
124 Aigner et al.
(16) as well as a reaction to actual culture conditions, e.g., maintenance of

cells in a low-density monolayer is known to destabilize the chondrocytic phe-
notype (17,18), whereas alginate beads have been suggested to preserve the
phenotype (19). Overall, different in vivo and in vitro model systems revealed
rather variable and partly contradicting results. Thus, realistically it will be
more appropriate to aim for in vivo and in vitro models that reflect defined
disease processes, rather than one model that addresses the disease as a whole.
c. Functional gene validation. Final confirmation that a gene product has a spe-
cific functional role in a cellular or disease process will have to come from
additional experimental work using overexpression or knockout/antisense tech-
nologies, to name a few. One way to limit the requirement for multiple tech-
niques to well-selected targets is to analyze and validate gene networks
involved in one regulatory context. For osteoarthritis research, one might relate
genes to major concepts of cartilage degeneration as follows: (1) hypercatabolic
activity of chondrocytes; (2) hypoanabolic activity of chondrocytes, as well as
(3) phenotypic alterations of chondrocytes within the tissue. All three concepts
have major potential for understanding the degeneration process and can pro-
vide a coordinate system in order to establish validation assays and to under-
stand the role of genes in disease.
3.4.2. Conclusion
There are constitutive drawbacks to gene expression technologies, as they do
not provide information on the relevance of factors such as (1) mRNA levels vs
protein levels, (2) the effects of posttranslational modifications of proteins, like
phosphorylation and dephosphorylation, (3) mRNA/protein degradation, and (4)
protein subcellular location. Other technical problems exist such as signal reli-
ability and reproducibility of the assay as well the availability of prototypical
tools in biostatistics and bioinformatics. Regardless of these drawbacks,
microarray systems and other technologies are initiating new research strategies
to aid in understanding disease pathogenesis as well as the development of drug
targets. Undoubtedly, it should be emphasized that most of the results obtained
to date should be considered somewhat preliminary, as technical tools and inter-
pretation approaches require further validation. The major challenge will be to
extract important information and knowledge from an obtained mass of data that
will be a basis for a new world of understanding the biological processes in
tissues and disease such as cartilage and arthritis (3,20).
4. Notes
1. Milling the cartilage prior to resuspension in lysis buffer dramatically increases
the efficiency of RNA isolation. In our laboratory, we use a 6800 Freezer/Mill
(SPEX CertiPrep) to grind tissue. Unlike other laboratory mills, a freezer mill is
an impact grinder that uses a steel impactor that moves back and forth between
the metal ends of a vial by strong magnetic fields. The liquid nitrogen freezes the
tissue sample, thus preventing RNA degradation. Since mechanical stress (cre-
ated when the tissue is milled between the end plugs and the impactor) results in
an increase in temperature, the cooling time between each cycle of grinding is
therefore necessary to preserve the RNA. In addition, it is also necessary to cool
the magnetic coils that generate the magnetic field.
2. The recommended protocol states that the samples should be centrifuged through
the column for 5 min. In our experience, after 5 min of centrifugation, the flow
through the column may become blocked, thus severely hindering sample process-
ing. The sequential application and centrifugation of the cleared lysate for only 1
min did not affect the flow of the cleared lysate through the column membrane.
3. Estimation of RNA quality is not trivial, and evaluation criteria also depends on
the source of cells or tissue used as the starting material. Thus, RNA isolation
from cells is usually straightforward and yields high amounts of high-quality
RNA (OD260/OD280 of about 2). RNA isolated from articular cartilage requires
some compromise and is usually not as abundant as that obtained from cells
(approx 2–10 µg RNA/g tissue) (2). In addition, the OD260/OD280 ratio of RNA
from tissue is lower than that isolated from cells (around 1.6–1.8) owing to the
presence of “contaminating” extracellular matrix components in the tissue.
4. The hybridization efficiency and specificity of two complementary nucleic acid
strands depends on many factors. These include sequence-specific parameters
such as length and GC content of the probe and target. Other parameters include
probe/target concentration, cation concentration, valency, pH, dielectric and
chaotropic media, incubation time, and incubation temperature, as well as sur-
face characteristics of the solid support and density spacing of the probe mol-
ecules synthesized on the surface of the membrane. Therefore, the hybridization
and posthybridization conditions have to be determined for each specific
microarray to achieve the best sensitivity and noise-level ratio. In practice, this
should be carried out by the manufacturer.
5. The Affymetrix Microarray Suite (MAS) software controls the fluidics station and
the scanner and also analyses the hybridization intensity data from the probe
arrays. MAS can then calculate a set of metrics that describe the probe set perfor-
mance. Whereas MAS 4.0 employed an empirical algorithm, the most recent ver-
sion, MAS 5.0, makes use of statistical algorithms and also generates detection
p-values, thus eliminating biologically irrelevant, negative expression intensities.
Acknowledgments
We are grateful to Dr. Audrey McAlinden for professional editing of the
manuscript. This work was supported by the by the IZKF (grant D4) and the
German Ministry of Research (grant 01GG9824).
References
1.
1 Eisen, M. B., Spellman, P. T., Brown, P. O., and Botstein, D. (1998) Cluster analy-
sis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA
95, 14,863–14,868.
126 Aigner et al.
2.
2 McKenna, L. A., Gehrsitz, A., Soeder, S., Eger, W., Kirchner, T., and Aigner, T.
(2000) Effective isolation of high quality total RNA from human adult articular
cartilage. Anal. Biochem. 286, 80–85.
3.
3 Grant, E. P., Pickard, M. D., Briskin, M. J., and Gutierrez-Ramos, J. C. (2002)
Gene expression profiles: creating new perspectives in arthritis research. Arthritis
Rheum. 46, 874–884.
4.
4 Lockhart, D. J. and Winzeler, E. A. (2000) Genomics, gene expression and DNA
array. Nature 405, 827–836.
5 Quackenbush, J. (2001) Computational analysis of microarray data. Nat. Rev.
5.
Genet. 2, 418–427.
6.
6 Schulze, A. and Downward, J. (2001) Navigating gene expression using
microarrays—a technology review. Nat. Cell Biol. 3, E190–E195.
7.
7 Lipshutz, R. J., Fodor, S. P., Gingeras, T. R., and Lockhart, D. J. (1999) High
density synthetic oligonucleotide arrays. Nat. Genet. 21, 20–24.
8.
8 Lockhart, D. J., Dong, H., Byrne, M. C., et al. (1996) Expression monitoring by
hybridization to high-density oligonucleotide arrays. Nat. Biotechnol. 14, 1675–1680.
9 Eberwine, J., Yeh, H., Miyashiro, K., et al. (1992) Analysis of gene expression in
9.
single live neurons. Proc. Natl. Acad. Sci. USA 89, 3010–3014.
10.
10 Velculescu, V. E., Madden, S. L., Zhang, L., et al. (1999) Analysis of human
transcriptomes. Nat. Genet. 23, 387–388.
11.
11 Zien, A., Aigner, T., Zimmer, R., and Lengauer, T. (2001) Centralization: a new
paradigm for the normalization of gene expression data. Bioinformatics 17,
S323–S331.
12.
12 Ben Dor, A., Bruhn, L., Friedman, N., Nachman, I., Schummer, M., and Yakhini,
Z. (2000) Tissue classification with gene expression profiles. J. Comput. Biol. 7,
559–583.
13 Manduchi, E., Grant, G. R., McKenzie, S. E., Overton, G. C., Surrey, S., and
13.
Stoeckert, C. J., Jr. (2000) Generation of patterns from gene expression data by
assigning confidence to differentially expressed genes. Bioinformatics 16, 685–698.
14.
14 Benjamini, Y., Drai, D., Elmer, G., Kafkafi, N., and Golani, I. (2001) Control-
ling the false discovery rate in behavior genetics research. Behav. Brain Res.
125, 279–284.
15.
15 Erickson, G. R., Gimble, J. M., Franklin, D. M., Rice, H. E., Awad, H., and Guilak,
F. (2002) Chondrogenic potential of adipose tissue-derived stromal cells in vitro
and in vivo. Biochem. Biophys. Res. Commun. 290, 763–769.
16.
16 Goldring, M. B., Birkhead, J. R., Sandell, L. J., Kimura, T., and Krane, S. M.
(1988) Interleukin 1 suppresses expression of cartilage-specific types II and IX
collagens and increases types I and III collagens in human chondrocytes. J. Clin.
Invest. 82, 2026–2037.
17 Benya, P. D. and Shaffer, J. D. (1982) Dedifferentiated chondrocytes reexpress the
17.
differentiated collagen phenotype when cultured in agarose gels. Cell 30, 215–224.
18.
18 von der Mark, K., Gauss, V., von der Mark, H., and Müller, P. K. (1977) Relation-
ship between cell shape and type of collagen synthesized as chondrocytes lose
their cartilage phenotype in culture. Nature 267, 531–532.
19.
19 Kolettas, E., Buluwela, L., Bayliss, M. T., and Muir, H. I. (1995) Expression of
cartilage-specific molecules is retained on long- term culture of human articular
chondrocytes. J. Cell Sci. 108, 1991–1999.
20. Attur, M. G., Dave, M. N., Tsunoyama, K., et al. (2002) “A system biology”
approach to bioinformatics and functional genomics in complex human diseases:
arthritis. Curr. Issues Mol. Biol. 4, 129–146.
21. Aigner, T., Zien, A., Gehrsitz, A., Gebhard, P. M., and McKenna, L. A. (2001)
Anabolic and catabolic gene expression pattern analysis in normal versus osteoar-
thritic cartilage using complementary DNA-array technology. Arthritis Rheum.
44, 2777–2789.
Nonviral Chondrocyte Transfection 129
11
High-Efficiency Nonviral
Transfection of Primary Chondrocytes
Jean F. Welter, Luis A. Solchaga, and Matthew C. Stewart
Summary
The introduction of foreign DNA into mammalian cells is an essential investigative tool
in molecular biology. Nonviral approaches to transfection offer the advantage of relatively
simple vector design, production, and purification and, for tissue engineering applications,
avoid many of the potential risks associated with virus-mediated transfection methods.
Unfortunately, primary cells, and in particular chondrocytes, are notoriously refractory to
conventional transfection approaches, and optimized transfection efficiencies in these cells
are extremely low (1–1.5%). In this chapter, we present three protocols that have proved
useful in transfecting primary chondrocytes at high efficiency (~70%). The first uses
radiofrequency electroporation, a transfection method that frequently works extremely well
in cell types that are difficult to transfect. It should be noted that electroporation is not
limited to DNA but that essentially any molecule can be introduced into the cell using this
approach. In addition to the primary protocol, we present two additional reliable, albeit less
efficient backup protocols, the first using exponential decay electroporation and the second
FuGENE™ 6 transfection.
Key Words: Chondrocytes; transfection; nonviral transfection; electroporation; radio fre-

quency electroporation; exponential decay electroporation; FuGENE™ 6 transfection.
1. Introduction
The introduction of foreign DNA into mammalian cells is an essential
investigative tool in molecular biology. It is used analytically for the study of
gene promoter function and functional testing of proteins, and has the poten-
tial to be used therapeutically to augment tissue engineering endeavors.
Nonviral approaches to transfection offer the advantage of relatively simple
vector design, production, and purification, and, for tissue engineering appli-
129
130 Welter, Solchaga, and Stewart
cations, avoidance of many of the potential risks associated with virus-medi-

ated transfection methods (1).
Highly efficient methods for transfecting established cell lines are used rou-
tinely in the laboratory; commercially available, off-the-shelf systems easily
achieve transfection efficiencies on the order of 70% for many established cell
lines. Unfortunately, some cell types, in particular primary cells, respond poorly
to these approaches. Chondrocytes are notoriously recalcitrant, and optimized
transfection efficiencies in these cells are on the order of only 1–5% using con-
ventional approaches (2). Higher efficiencies have been reported using more
complex delivery methods (3,4).
Electroporation is a transfection approach that frequently works well
for cell types that are refractory to other methods of transfection. In this
approach, cells are subjected to a brief electric field pulse of appropriate
waveform, amplitude, and duration (5–8). This exposure results in the tran-
sient permeabilization of the cytoplasmic membrane, which allows extracel-
lular molecules to enter the cell. However, a subset of the cells does not
recover from the membrane breakdown, and a high rate of cell death may
occur (9). Depending on the subsequent disposition of the cells, the survival
rate may become important. For example, to initiate pellet (10) or micromass
(11) chondrogenesis, a minimum concentration of viable cells is required.
Usually, cells are electroporated in suspension in a specialized cuvet that
incorporates a pair of planar electrodes. However, it can be applied to adher-
ent cells or even used in vivo (12–18). Furthermore, it should be noted that
electroporation is not limited to DNA, but that essentially any molecule can
be introduced into the cell using this approach (8).
An electroporation variant, known as radiofrequency (RF) electroporation,
which combines squarewave and radiofrequency (5–50 kHz) AC pulses to trans-
fect the cells, was introduced by Donald Chang in the late 1980s (19,20). The
primary advantages of this method over conventional (exponential decay)
electroporation are equal or better transfection efficiencies, lower cell death,
and lower DNA requirements (21,22). Transfection efficiencies of better than
30% can be achieved routinely, and with cell type-specific optimization and
careful handling of the cells, the efficiency can exceed 70% of the surviving
cells. This system has proved extremely valuable for transfecting otherwise
difficult cell types.
Our primary approach to high-efficiency transfection of primary
chondrocytes utilizes RF electroporation and is discussed in this chapter. We
use a commercial system, the Bio-Rad Gene Pulser II with the RF module. In
addition to the primary protocol, we present two additional reliable, albeit less
efficient backup protocols, the first using exponential decay electroporation,
and the second, FuGENE™ 6 transfection.
2. Materials
2.1. Primary Protocol: RF Electroporation
1. A Bio-Rad Gene Pulser II with RF module or equivalent device (Bio-Rad cat. no.
165-2105 and 165-2112; see Note 1).
2. Sterile electroporation cuvets: 2-mm electrode gap electroporation cuvets (Bio-
Rad cat. no. 165-2086, or similar).
3. Electroporation buffer: 272 mM sucrose, 7 mM sodium phosphate, pH 7.4, 1 mM
MgCl2. Prepare the solution, and then filter sterilize using a 0.2-µm filter. Store
frozen aliquots at –20°C.
4. Chondrocytes.
5. Plasmid DNA.
6. Complete growth medium without antibiotics (see Note 2): 0.25% trypsin
(Invitrogen Life Technologies, Carlsbad, CA, cat. no. 15050057), calcium- and
magnesium-free Hanks’ balanced salt solution (HBSS; Invitrogen Life Technolo-
gies, CA, cat. no. 14170112), sterile T10E1 (10 mM Tris HCl, 1 mM EDTA in
sterile water, pH 8.0) or water.
7. Standard tissue culture equipment and supplies: incubator, cell culture hood,
inverted microscope, hemacytometer, tissue culture plasticware, sterile Pasteur
pipets (any supplier).
2.2. Backup Protocol 1: Exponential Decay Electroporation

1. A Bio-Rad Gene Pulser II with the Capacitance Extender II and (optionally)
Pulse Controller Plus modules (Bio-Rad cat. no. 165-2105, 165-2107, and
165-2110)
2. Sterile 4-mm electrode gap electroporation cuvets (Bio-Rad cat. no. 165-2088, or
similar).
3. Electroporation buffer: Ca2+ and Mg2+-free Tyrode’s solution supplemented with
5% calf serum.
4. Chondrocytes, plasmid DNA, growth medium, trypsin, HBSS, sterile T10E1 or
water, standard equipment and supplies: see Subheading 2.1., items 4–7.
2.3. Backup Protocol 2: FuGENE 6 Transfection

1. FuGENE 6 Transfection Reagent (Roche Diagnostics, Indianapolis, IN, cat. no.
P/N 1 814 443).
2. Chondrocytes, plasmid DNA, standard equipment and supplies: see Subheading
2.1., items 4, 5, and 7.
3. Opti-MEM I® reduced serum medium (Invitrogen Life Technologies, cat. no.
31985070).
4. Sterile 1.5-mL Eppendorf tubes.
5. Ultra Low Attachment plates (Corning Costar, Acton, MA, cat. no. 3473)
Fig. 1. Schematic of principal components of an radio frequency (RF) electro-

poration burst train. Two representative bursts are shown.
3. Methods
3.1. Primary Protocol: RF Electroporation
In this approach, the cells are exposed to an electrical field that varies over
time in a complex, preprogrammed waveform, or burst, resulting from the super-
position of a RF and a DC pulse of the same width. Trains of identical bursts can
be delivered. The critical parameters are illustrated in Fig. 1, and include total
voltage, RF frequency,% modulation, and burst duration, number, and interval.
This approach works best in low conductivity buffers.
3.1.1. Planning
Determine the total number of transfections required for the experiment.
Usually, we use 3 × 105 to 1 × 106 cells per transfection, with 6 × 105 working
best for human chondrocytes.
3.1.2. Setting up
3.1.2.1. GENERAL
Wipe down the work area in the cell culture hood with 70% ethanol. Prepare
and label a sterile conical tube for each cell-plasmid combination to be tested,
and the required number of cell culture plates to receive the transfected cells.
Time requirements will, of course, vary with the scale of the experiment, but
they are minimal. In general, the apparatus can be set up in 5 min. Trypsinizing,
washing, and counting the cells can be accomplished in 45 min, and the actual
electroporation step takes about 1 min for each cuvet.
Fig. 2. Rear panel electrical connections for RF electroporation using the Bio-Rad
Gene Pulser II and RF modules.
3.1.2.2. SETTING UP THE ELECTROPORATOR

(These instructions refer to the Bio-Rad Gene Pulser II equipment. Refer to
the appropriate user’s manual if you are using a different unit.)
1. Place the electroporation apparatus in a corner of the cell culture hood.
2. At the back of the machine, connect the 9-pin D connector from the RF module
port labeled Gene Pulser II Rear to the base unit port labeled Accessory. This
should be the only D connector cord connected to the base unit.
3. Connect the high voltage leads (thick black coiled cord) from the RF module to
the socket labeled Capacitance Extender on the base module. This should be the
only coiled cord connected to the base unit.
4. Plug both power cords into grounded outlets. When done, the setup should look
like Fig. 2.
5. Connect cuvet holder to the RF module, not to the base module. Set the rotary
switch on the base unit to High Cap. Now, turn on the electroporator base unit
and RF modules.
3.1.2.3. PROGRAMMING THE ELECTROPORATOR
1. On the RF unit, make sure the black knob is pointing to any position other than
OFF, and then press both red buttons on the base unit simultaneously. The base
unit is now in RF mode, and the display should read -rf- (Fig. 3).
2. Program the electroporator to run the desired parameters using the front panel
controls on the RF module (Fig. 3). We have optimized the following parameters
for human chondrocytes:
Fig. 3. Front panel of the RF module. Rotary function selector at right, and touch keys
at left. Note high-voltage connection to the electroporation cuvet at the lower right. This
connection should not be made to the similar set of output jacks on the base unit.
Total Volts 25 V (125 V/cm)

% Modulation 100
RF frequency 30 kHz
Burst duration 2 ms
No. of bursts 10
Burst interval 0.1 s
3. Program each parameter in turn by selecting it with the Function Select knob and
using the Raise and Lower buttons to adjust the setting to the desired value (Fig. 3).
When done, save the program, using the Save Program buttons on the front panel
(see also Note 3).
3.1.3. DNA
The DNA should be highly purified. We have found Qiagen column purifi-
cation to be adequate. The DNA should be washed thoroughly with 70% etha-
nol to remove residual salts.
1. Resuspend the DNA in a sterile, low ionic strength vehicle (T10E1 or water), at
high concentration (1 µg/µL or greater) to minimize the volume contribution of
the plasmid to the overall transfection mix. Ten micrograms per transfection
works well, but considerably less will also yield positive transfections. The
amount of DNA will impact on the transfection efficiency, so it is important to
maintain a consistent concentration between experiments.
2. Heat the plasmid DNA solution to 70°C for 5–10 min, vortex briefly, and then,
using sterile technique, deposit the plasmid volume required at the bottom of a
sterile conical tube prepared as in Subheading 3.1.2.1.
3.1.4. Cells
For human chondrocytes, 6 × 105 cells per electroporation work best. One can
expect approx 70% survival at the recommended parameters. Cells that are
subconfluent just before the electroporation seem to survive the process better.
The cells should be treated gently: prolonged trypsinization, vigorous pipeting,
high g-forces during centrifugation, and prolonged dwell time in the electropor-
ation buffer all have a negative impact on transfection efficiency and viability.
1. Trypsinize the cells after rinsing with HBSS. Check the plates every few min, and
stop the trypsinization using complete medium once the cells are free floating.
2. Centrifuge the cell suspension in 50-mL conical tubes (250g, 6 min) and remove
the supernatant. Resuspend the cells in the electroporation buffer.
3. Centrifuge (250g, 6 min) and remove the supernatant. Repeat this washing step
once, and then resuspend the cells in electroporation buffer.
4. Count the cells using a hemacytometer and adjust the volume with electroporation
buffer so that there are 1.5 × 106 cells/mL (6 × 105 cells per 400 µL—the volume
between the cuvet electrodes in a 2-mm cuvet).
3.1.5. Electroporation
Select the desired electroporation program (see Subheading 3.1.2.3.). To
switch between stored programs, adjust the program number using up and down
arrows, and then press the two buttons labeled USE PROGRAM simulta-
neously. Unn will be displayed, with nn being the number of the program now
in use. The following steps should be completed quickly, so the cells do not
have time to sediment.
1. Gently mix the required volume of cells with the DNA in the labeled conical tubes.
2. Transfer 400 µL of the cell/DNA mixture to an electroporation cuvet, and place
the cuvet in the slider of the electroporation cuvet holder. The cuvet is keyed
and will fit easily in one orientation only. Push the slider into the holder until it
engages with a click.
3. Press and hold both red buttons on the RF unit simultaneously until a beep sounds.
Release the buttons.
3.1.6. Post Electroporation

Best results are achieved if the time between trypsinization of the cells and
their return to normal medium after the electroporation is kept at a minimum
(see Note 4).
1. Immediately remove the cuvet from the holder and add 1 mL of antibiotic-free
medium to the cuvet.
2. Gently aspirate the cell suspension from the cuvet using a sterile pipette and plate
it in the desired final volume of antibiotic-free medium.
3. Return the plated cells to the cell culture incubator. If stable tranfectants are
desired, selection can begin after 24–48 h (see Note 5).
Fig. 4. Schematic of principal components of an exponential decay electroporation

pulse. V0 = initial voltage, τ = time constant = time for voltage to decay from V0 to V0/e.
3.2. Backup Protocol 1: Exponential Decay Electroporation

In this approach, the cells are exposed to a single pulse which decays expo-
nentially through time. The pulse is defined by two pulse parameters, the ini-
tial field strength (kV/cm) and the time constant (τ) (Fig. 4). Both of these
must be optimized for each cell type. The field strength is the initial voltage V0
input into electroporator divided by the distance between the cuvet electrodes.
The time constant τ is equivalent to the amount of time it takes for V0 to drop to
V0/e (i.e., about 37% of V0). τ is also the product of the total resistance and
capacitance of the pulse circuit (τ = R × C) in ms, Ω, µF, respectively. It is thus
dependent on the conductivity and the amount of buffer between the electrodes.
3.2.1. Planning (see Subheading 3.1.1.)
3.2.2. Setting Up
3.2.2.1. SETTING UP THE ELECTROPORATOR
1. Place the electroporation apparatus in a corner of the cell culture hood.
2. At the back of the unit, optionally connect the high-voltage leads (thick black
coiled cord) from the Pulse Controller module to the socket labeled Pulse Con-
troller on the base module.
3. Connect the high-voltage leads (thick black coiled cord) from the Capacitance
Extender II module to the socket labeled Capacitance Extender on the base mod-
Fig. 5. Rear panel electrical connections for exponential decay electroporation using
the Bio-Rad Gene Pulser II and Capacitance Extender and Pulse Controller modules.
ule. Connect the 25-pin D connector on the Capacitance Extender module to the
25-pin D connector on the base unit.
4. Plug the power cord into a grounded outlet. When done the setup should look like
Fig. 5.
5. Connect the cuvet holder to the base module. Now, turn on the electroporator
base unit.
3.2.2.2. PROGRAMMING THE ELECTROPORATOR

The electroporator must be programmed to run the desired parameters using
the front panel controls. Set the rotary control to High Cap to connect the
capacitance extender module (Fig. 6). If connected, set the knobs on the pulse
Controller Plus to High Range and ∞. For human chondrocytes, the following
parameters work well: 0.400 kV, 250 µF (base unit), and ∞Ω (Pulse Controller
Plus in open circuit mode; see Note 6). The parameters are entered using the
front panel buttons.
3.2.3. DNA
The DNA should be prepared as described in Subheading 3.1.3. Exponential
decay electroporation requires large amounts of DNA for maximum efficiency;
Fig. 6. Front Panel view of the exponential decay electroporation setup using the
Bio-Rad Gene Pulser II and Capacitance Extender and Pulse Controller modules. The
latter is in open circuit (∞Ω), and the high-voltage connection to the electroporation
cuvet is made to the base unit.
40 µg per transfection works well, but considerably less will also yield positive
transfections. Heat the plasmid DNA solution to 70°C for 5–10 min, and then,
using sterile technique, deposit the total volume of plasmid required at the bot-
tom of a sterile conical tube prepared as in Subheading 3.2.2.1.
3.2.4. Cells
Some 3 × 105 to 1 × 106 cells can be transfected at a time, with 6 × 105
working best for human chondrocytes. Exponential decay electroporation kills
many more cells than RF electroporation. Expect only about 30% survival at
the recommended parameters. Cells that were subconfluent seem to survive
the electroporation process better. As described in Subheading 3.1.4., the cells
should be treated gently to enhance viability. Additionally, keeping the cells
on ice before and immediately after electroporation improves viability.
The cells should be prepared as described in Subheading 3.1.4. with the

final resuspension being in cold electroporation buffer. Count the cells using a
hemacytometer and adjust the volume with cold electroporation buffer so that
there are 0.75 × 106 cells/mL.
3.2.5. Electroporation
1. Gently mix the required volume of cells with the DNA in the labeled conical tubes.
2. Transfer 800 µL of the cell/DNA mix to a sterile cuvet and place the cuvet in the
slider of the electroporation cuvet holder. The cuvet is keyed and will fit easily in
one orientation only. Push the slider into the holder until it engages with a click.
3. Press and hold both red pulse buttons simultaneously until the beep sounds (CHg
will flash and then PLS will be displayed during this process).
4. Release both pulse buttons after the beep. The time constant is automatically
displayed after every pulse. Record it, as well as the actual volts and sample
resistance (see Note 7). This information can be useful for troubleshooting and
for verifying interexperimental consistency.
3.2.6. Post Electroporation

Results improve if the time between trypsinization of the cells and their return
to normal medium after the electroporation is kept at a minimum (see Note 4).
However, the cells are fragile immediately after electroporation.
1. Remove the cuvet from the holder and add 1 mL of cold, antibiotic-free medium
to the cuvet.
2. Place the cuvet on ice and wait for 10–15 min (7), before gently aspirating the
cell suspension from the cuvet using a sterile pipette. It has been reported that
reheating the cells to 37°C instead of keeping them on ice at this stage improves
transfection efficiency (23), but we have not verified this in our system.
3. Plate the cells in the desired final volume of antibiotic-free medium, and return
the plates to the incubator.
3.3. Backup Protocol 2: FuGENE 6 Transfection

This protocol was developed for the transfection of chondrocytes maintained
under nonadherent culture conditions; however the procedure is also appli-
cable to monolayer cultures. FuGENE 6 is a proprietary, lipid-based transfec-
tion reagent that has been shown to provide significantly higher transfection
efficiencies in chondrocytes than similar cationic liposome-based alternatives
in two comparative analyses (3,24). The transfection efficiency using this pro-
tocol is generally 10–40%. This is not sufficient to assess the effects of specific
gene expression on the entire target cell populations without selection, owing
to the high negative-cell background. It is adequate for promoter/reporter-based
experiments, establishing stable transfectants and creating positive control
lysates and media.
Fig. 7. Effect of time after isolation on chondrocyte transfection efficiency. Pri-

mary chondrocytes were transfected with a cytomegalovirus (CMV) promoter-driven
luciferase reporter at varying times after isolation from cartilage. Luciferase activities,
indicated in relative light units (RLU), were measured 48 h after transfection (n = 3).
3.3.1. Setting Up
Wipe down the work area in the cell culture hood with 70% ethanol. Prepare
and label sterile tubes for each plasmid combination and the required number
of cell culture plates to receive the transfected cells. The chondrocytes should
be transfected immediately after collagenase isolation to maximize efficiency
(Fig. 7). Therefore, the FuGENE 6/DNA complex should be set up concur-
rently with the final stages of the cell isolation procedure.
3.3.2. DNA
The DNA should be highly purified. We have found Qiagen column purifica-
tion to be adequate. The DNA should be washed thoroughly with 70% ethanol to
remove residual salts. Resuspend the DNA at high concentration (1 µg/µL or
greater) to minimize volume contribution, in a sterile vehicle (T10E1 or water).
3.3.3. Cells
The transfection efficiency of primary chondrocytes falls substantially with
time after isolation (Fig. 7). Therefore, optimal results will be obtained using
chondrocytes immediately after being released from cartilage by collagenase
digestion (see Note 8). For reporter-based experiments, we generally aliquot
chondrocytes into groups of 2–3 × 105 cells, with each group maintained in a
2.0-cm2 nonadherent well.
The following protocol details the steps required to transfect a single well of
3 × 105 cells. The quantities and volumes can be scaled up as necessary, to
accommodate larger populations of target cells or to prepare sufficient reagents
for replicates. It is advisable to prepare sufficient complex solution for an extra
well, to ensure adequate quantities for the entire experimental group.
1. Place 97 µL of Opti-MEM in a 1.5-mL Eppendorf tube.
2. Add 3 µL of FuGENE 6 reagent directly into the medium, so as to prevent the
FuGENE 6 from directly contacting the plastic surface of the tube.
3. Mix gently, and then add 1 µg of DNA (usually in 1–2 µL) to the Opti-MEM
containing the FuGENE 6.
4. Mix by gently tapping the tube, and incubate at room temperature for 15–30 min.
5. Chondrocyte isolation from the digestion buffer can be completed during this
interval. After isolation, washing, and counting, place 3 × 105 cells in 500 µL of
Opti-MEM I into a 2-cm-diameter, ultra-low attachment well. Do not supple-
ment the medium with ascorbic acid until the transfection has been carried out.
6. Add the FuGENE 6/DNA complex to the well and swirl the dish to disperse the
complex and cells evenly.
7. Return the cells to an incubator and maintain in the FuGENE 6/DNA medium for
at least 3 h.
3.3.4. Post Transfection

The cultures can be supplemented with appropriate medium after 3 h. It is not
necessary to remove the medium containing the FuGENE 6/DNA complex. We
routinely maintain chondrocyte cultures in serum-free Opti-MEM, so this only
entails adding a further 500 µL of medium, with the required ascorbic acid
supplementation. Using this protocol, we have found that constitutive luciferase
expression is maximal 48–72 h after transfection and persists at detectable (24-h
equivalent) levels for at least 9 d (Fig. 8). These data were obtained from
chondrocytes maintained in serum-free Opti-MEM. Maintenance in serum-
supplemented medium may alter these parameters.
4. Notes
1. Unfortunately, although (or perhaps because) the Gene Pulser II RF is an extremely
versatile, configurable, and thus complex system, this particular model was discon-
tinued by Bio-Rad in 2002. Many of them are still in operation, however, and they
can occasionally be found on the used equipment market. It has been replaced by a
less versatile square-wave capable electroporation system (Gene Pulser Xcell); this
and other devices, e.g., from BTX, or Amaxa (2), may be able to produce similar
transfection efficiencies (25). Alternately, a competent electronics shop may be
able to produce a device with capabilities similar to those of the commercial RF
system (19).
2. Deleting antibiotics from the post transfection medium improves cell survival.
Fig. 8. Persistence of luciferase activity in chondrocyte aggregates. Luciferase activ-

ity, indicated in relative light units (RLU), was assessed at varying time points after
transfection of a cytomegalovirus (CMV) promoter-driven reporter in chondrocytes
maintained in serum-free medium (n = 3).
3. Optimizing RF electroporation: the parameters presented work well for cul-

tured primary human chondrocytes (and, incidentally, for human bone mar-
row-derived mesenchymal stem cells). Other parameters may work better for
passaged chondrocytes or for chondrocytes from different species. The follow-
ing guidelines may be useful in optimizing RF electroporations.
a. We have found that 30 kHz 100% modulation appears to work well for most
cell types.
b. Also, in general, both transfection efficiency and cell death rise with the num-
ber of bursts, so an optimal tradeoff will have to be found empirically.
c. Focusing on optimizing the Total Voltage and Burst Durations seems to be
the most expedient approach. Assuming a 2 mm cuvet, a good starting point
would be setting up an experimental grid of nine conditions, e.g.,
Burst duration Total volts
(ms) 25 50 100
0.5 1 2 3
1 4 5 6
2 7 8 9
while maintaining the RF frequency at 30 kHz, percent modulation at 100%,
burst number at 5 and burst interval at 1 s. Pick the condition that results in
the highest transfection efficiency, and then increase the burst number.
Preliminary optimization is most conveniently done by transfecting with a
luciferase reporter vector. Assessing transfection efficiency can be conveniently
done using a green fluorescent protein (GFP) reporter vector, e.g., pEGFP-N1
(Clontech, BD Biosciences, Franklin Lakes, NJ, http://www.clontech.com/

techinfo/vectors/vectorsE/pEGFP-N1.shtml). The transfected cells are plated
onto chamber slides, e.g., Nunc™ Lab-Tek™ (Nalge Nunc International, Roch-
ester, NY, cat. no. 155380,). Percent efficiency is determined by dividing the
number of GFP-positive cells by the total number of cells counted on several
random 10× fields, × 100. Peak expression of GFP usually occurs on the third
day after transfection. A bacterial β-galactosidase reporter system followed by
X-gal detection can also be used; however, the sensitivity of this assay is sev-
eral orders of magnitude lower than that of either luciferase- or GFP-based sys-
tems.
4. Leaving the cells in the electroporation buffer for an extended period decreases
both transfection efficiency and viability. Therefore, complete all the electro-
porator setup and programming steps, quantify the DNA, label the culture
dishes, and so on, before harvesting the cells. If a large number of transfections
are planned, consider breaking up the experiment into smaller groups. It may be
helpful to have an assistant handle the postelectroporation plating in another
cell culture hood.
5. Stable transfections: both electroporation protocols yield higher numbers of
stable transformants than one would expect using conventional transfection
techniques. Selection of stable transformants can begin the day after electro-
poration. From the literature, selection levels for human chondrocytes trans-
fected with the neomycin resistance marker are in the range of 350–400 µg/
mL of G418 (26).
6. The parameters given in Subheading 3.2.2.3. work well for human
chondrocytes. Other species will probable require further optimization. Cell-
specific optimization is important if for no other reason than that the required
field strength depends on the diameter of the cell (27). Setting up an optimi-
zation grid varying V0 up and down around 400 V, while maintaining the
capacitance, should prove to be a good starting point. Careful washing of the
cells in electroporation buffer, and use of the Pulse-Track system (see Note 7)
will help ensure reproducible results.
7. The Gene Pulser II unit incorporates a system called Pulse-Track that monitors
the resistance of the sample and delivers the desired voltage regardless of
sample volume or conductivity. Using this system according to the manu-
facturer’s instructions can improve the reproducibility of exponential decay
electroporations.
8. When transfection with FuGENE 6 cannot be carried out immediately after
chondrocyte isolation, the effects of newly synthesized pericellular matrix
can be mitigated by pretreating the chondrocytes with 4 U/mL hyaluronidase
for 6 h prior to and during the transfection (3).
References
1.
1 Reid, T., Warren, R., and Kirn, D. (2002) Intravascular adenoviral agents in can-
cer patients: Lessons from clinical trials. Cancer Gene Ther. 9, 979–986.
2.
2 Hamm, A., Krott, N., Breibach, I., Blindt, R., and Bosserhoff, A. K. (2002) Effi-
cient transfection method for primary cells. Tissue Eng. 8, 235–245.
3.
3 Stove, J., Fiedler, J., Huch, K., Gunther, K. P., Puhl, W., and Brenner, R. (2002)
Lipofection of rabbit chondrocytes and long lasting expression of a lacZ reporter
system in alginate beads. Osteoarthritis Cartilage 10, 212–217.
4.
4 Goomer, R. S., Deftos, L. J., Terkeltaub, R., et al. (2001) High-efficiency non-
viral transfection of primary chondrocytes and perichondrial cells for ex-vivo
gene therapy to repair articular cartilage defects. Osteoarthritis Cartilage 9,
248–256.
5 Neumann, E., Schaefer-Ridder, M., Wang, Y., and Hofschneider, P. H. (1982)
5.
Gene transfer into mouse lyoma cells by electroporation in high electric fields.
EMBO J. 1, 841–845.
6.
6 Wong, T. K. and Neumann, E. (1982) Electric field mediated gene transfer.
Biochem. Biophys. Res. Commun. 107, 584–587.
7.
7 Potter, H., Weir, L., and Leder, P. (1984) Enhancer-dependent expression of human
kappa immunoglobulin genes introduced into mouse pre-B lymphocytes by electro-
poration. Proc. Natl. Acad. Sci. USA 81, 7161–7165.
8.
8 Gehl, J. (2003) Electroporation: theory and methods, perspectives for drug deliv-
ery, gene therapy and research. Acta Physiol. Scand. 177, 437–447.
9 Kekez, M. M., Savic, P., and Johnson, B. F. (1996) Contribution to the biophysics
9.
of the lethal effects of electric field on microorganisms. Biochim Biophys. Acta
1278, 79–88.
10.
10 Lennon, D. P., Haynesworth, S. E., Arm, D. M., Baber, M. A., and Caplan, A. I.
(2000) Dilution of human mesenchymal stem cells with dermal fibroblasts and
the effects on in vitro and in vivo osteochondrogenesis. Dev. Dyn. 219, 50–62.
11.
11 Ahrens, P. B., Solursh, M., and Reiter, R. S. (1977) Stage-related capacity for
limb chondrogenesis in cell culture. Dev. Biol. 60, 69–82.
12.
12 Raptis, L. and Firth, K. L. (1990) Electroporation of adherent cells in situ. DNA
Cell Biol. 9, 615–621.
13 Zheng, Q. A. and Chang, D. C. (1991) High-efficiency gene transfection by in
13.
situ electroporation of cultured cells. Biochim. Biophys. Acta 1088, 104–110.
14.
14 Wegener, J., Keese, C. R., and Giaever, I. (2002) Recovery of adherent cells after
in situ electroporation monitored electrically. Biotechniques 33, 348–357.
15.
15 Saito, T. and Nakatsuji, N. (2001) Efficient gene transfer into the embryonic
mouse brain using in vivo electroporation. Dev. Biol. 240, 237–246.
16.
16 Ohashi, S., Kubo, T., Kishida, T., et al. (2002) Successful genetic transduction in
vivo into synovium by means of electroporation. Biochem. Biophys. Res. Commun.
293, 1530–1535.
17
17. Kishimoto, K. N., Watanabe, Y., Nakamura, H., and Kokubun, S. (2002) Ectopic
bone formation by electroporatic transfer of bone morphogenetic protein-4 gene.
Bone 31, 340–347.
18.
18 Muramatsu, T., Nakamura, A., and Park, H. M. (1998) In vivo electroporation: a
powerful and convenient means of nonviral gene transfer to tissues of living ani-
mals (Review). Int. J. Mol. Med. 1, 55–62.
19.
19 Chang, D. C. (1989) Cell poration and cell fusion using an oscillating electric
field. Biophys. J. 56, 641–652.
20.
20 Chang, D. C., Gao, P. Q., and Maxwell, B. L. (1991) High efficiency gene trans-
fection by electroporation using a radio-frequency electric field. Biochim. Biophys.
Acta 1092, 153–160.
21.
21 Zald, P. B., Cotter, M. A., and Robertson, E. S. (2001) Strategy for increased
efficiency of transfection in human cell lines using radio frequency electropora-
tion. Prep. Biochem. Biotechnol. 31, 1–11.
22 Zald, P. B., Cotter, M. A., 2nd, and Robertson, E. S. (2000) Improved transfec-
22.
tion efficiency of 293 cells by radio frequency electroporation. Biotechniques
28, 418–420.
23.
23 Rols, M. P., Delteil, C., Serin, G., and Teissie, J. (1994) Temperature effects on
electrotransfection of mammalian cells. Nucleic Acids Res. 22, 540.
24.
24 Madry, H., and Trippel, S. B. (2000) Efficient lipid-mediated gene transfer to
articular chondrocytes. Gene Ther. 7, 286–291.
25.
25 Gehl, J., Skovsgaard, T., and Mir, L. M. (1998) Enhancement of cytotoxicity by
electropermeabilization: an improved method for screening drugs. Anticancer
Drugs 9, 319–325.
26 Piera-Velazquez, S., Jimenez, S. A., and Stokes, D. (2002) Increased life span of
26.
human osteoarthritic chondrocytes by exogenous expression of telomerase. Arthri-
tis Rheum. 46, 683–693.
27. Kotnik, T., Bobanovic, F., and Miklavcic, D. (1997) Sensitivity of transmembrane
voltage induced by applied electric fields—a theroetical analysis. Bioelectrochem.
Bioenergetics 43, 285–291.
In Vitro Gene Transfer to Articular Cells 147
12
In Vitro Gene Transfer to Chondrocytes
and Synovial Fibroblasts by Adenoviral Vectors
Jean-Noel Gouze, Martin J. Stoddart, Elvire Gouze, Glyn D. Palmer,

Steven C. Ghivizzani, Alan J. Grodzinsky, and Christopher H. Evans
Summary
The major requirement of a successful gene transfer is the efficient delivery of an exog-
enous therapeutic gene to the appropriate cell type with subsequent high or regulated levels of
expression. In this context, viral systems are more efficient than nonviral systems, giving higher
levels of gene expression for longer periods. For the application of osteoarthritis (OA), gene
products triggering anti-inflammatory or chondroprotective effects are of obvious therapeutic
utility. Thus, their cognate genes are candidates for use in the gene therapy of OA. In this
chapter, we describe the preparation, the use, and the effect of the transduction of chondrocytes
or synovial fibroblasts with an adenoviral vector encoding the cDNA for glutamine: fructose-
6-phosphate amidotransferase (GFAT). This is intended to serve as an example of a technology
that can be used to evaluate the biological effects of overexpression of other cDNAs.
Key Words: Gene transfer; adenovirus; GFAT; chondrocyte; synoviocyte; gene therapy;
osteoarthritis; interleukin-1; glucosamine.
1. Introduction
Osteoarthritis (OA) is the most common joint disorder and is associated with
a dramatically impaired quality of life (1). It is increasingly viewed as a dis-
ease of the entire joint, involving cartilage, the underlying bone, and the syn-
ovial lining of the joint capsule (2). Present treatments for OA remain
unsatisfactory. Although many pharmacologies diminish pain and inflamma-
tion, none is capable of blocking the progression of the disease (3,4).
Advances in molecular biology combined with a better understanding of the
basic biology of OA have opened new therapeutic opportunities for the treatment
of this disorder. Many of these potential new drugs with which to treat OA are
proteins; although they have potent activities, they are difficult to deliver to OA
joints in a manner that achieves sustained therapeutic intra-articular concentration.
147
148 Gouze et al.
Because genes are able to produce proteins locally for extended periods, their trans-
fer to specific cells within the joint, such as chondrocytes or synoviocytes, are
attractive alternatives to protein delivery in the context of treating OA.
To illustrate this point, the present chapter describes the construction and
utilization of an adenoviral vector containing the cDNA for glutamine:fructose-
6-phosphate amidotransferase (GFAT). Indeed, based on the recent attention
afforded by studies supporting the notion that glucosamine may have benefi-
cial effects in OA (5–11), we have developed an approach using GFAT as a
transgene. GFAT is the rate-limiting step of glucosamine synthesis within cells
(12); thus, by overexpressing the cDNA for GFAT in certain cells in the joint,
it may be possible to elevate the intra-articular levels of glucosamine and its
derivatives. Adenoviral vectors have been used to deliver the GFAT cDNA
successfully to synoviocytes cocultured with chondrocytes. In the presence
of interleukin-1 (IL-1), GFAT gene transfer was found to maintain matrix syn-
thesis by chondrocytes and also significantly reduced production of various
inflammatory mediators such as nitric oxide and prostaglandin E2 (13).
Because of their ease of use, efficiency of gene delivery, and high levels of
resulting transgene expression, adenoviral vectors have proved to be valuable
tools for evaluating the in vivo or in vitro activities of various transgene prod-
ucts (14–16). These vectors can be readily propagated to high titer and can
efficiently infect and transduce many cell types, including those of articular
tissues, such as synovial lining cells and chondrocytes. This chapter details the
infection of primary articular chondrocytes and synoviocytes with a recombi-
nant adenoviral vector encoding GFAT (Ad.GFAT). The process entails the
cloning of human GFAT cDNA, the generation and propagation of a novel
recombinant adenovirus, and the large-scale preparation of chondrocytes and/
or synoviocytes from joint tissues. Following infection of cell cultures, the
effect of GFAT overexpression can be demonstrated by protection against IL-1
challenge assessed by the changes in nitrate oxide (NO) levels.
This chapter describes methods by which the effects of GFAT
overexpression following gene transfer may be assessed and is intended to
serve as an example; the same technology can be used to evaluate the biologi-
cal effect of other cDNA overexpression.
2. Materials
1. TRIzol® reagent (Life Technologies).
2. Chloroform.
3. Isopropanol.
4. 70% Ethanol.
5. 95% Ethanol.
6. Dithiothreitol.
7. TE buffer: 10 mM Tris-HCl, pH 7.4, 1 mM ethylenediaminetetraacetic (EDTA).
8. RQ1 DNase (Promega).

9. Phenol/chloroform/isoamyl alcohol (25:24:1 v/v/v).
10. Reverse transcriptase-polymerase chain reaction (RT-PCR) reagents (Life Tech-
nologies, unless otherwise noted).
a. Random primers.
b. Murine Moloney leukemia virus (MMLV) RNase H– reverse transcriptase.
c. RNase OUT, ribonuclease inhibitor.
d. Deoxynucleotides (100 mM).
e. Forward and reverse oligodeoxynucleotides.
f. Vent DNA polymerase (New England Biolabs).
11. Reverse transcription buffer: 50 mM Tris-HCl, pH 8.4, 75 mM KCl, 3 mM MgCl2.
12. PCR buffer: 10 mM Tris-HCl, pH 9.0, 50 mM KCl, 1.5 mM MgCl2.
13. pAdlox adenoviral shuttle plasmid (Genebank accession number U62024; gener-
ous gift of Dr. Stephen Hardy; see ref. 17).
14. DH5α™ Competent Cells (Invitrogen).
15. Bacterial culture reagents (e.g., Sigma).
a. LB media: 1% tryptone, 0.5% yeast extract, 1% NaCl (w/v), pH 7.0.
b. LB agar: LB media + 15 g/L Bacto-agar.
c. Ampicillin (50 mg/mL).
16. QIAquick gel extraction kit (Qiagen).
17. Qiagen Tip 2500 plasmid kit (Qiagen).
18. T4 DNA ligase (Life Technologies).
19. EcoRI restriction endonuclease (Life Technologies).
20. BsaBI restriction endonuclease (New England Biolabs).
21. BamHI restriction endonuclease (Life Technologies).
22. Tissue culture reagents (e.g., Life Technologies, Sigma-Aldrich).
a. Ham’s F-12/Dulbecco’s modified eagle’s medium (DMEM).
b. DMEM.
c. Trypsin-EDTA: 0.05% trypsin, 0.53 mM EDTA.
d. Gey’s balanced salt solution (GBSS).
e. Phosphate-buffered saline (PBS).
f. Fetal bovine serum (FBS).
g. 10 mg/mL Penicillin /10,000 U/mL streptomycin in 0.85% saline.
h. Protease type XIV, from Streptomyces griseus.
i. Collagenase type II, from Clostridium histolyticum.
j. Dispase, from Bacillus polymyxa.
23. Tissue culture vessels (e.g., Becton Dickinson, NalgeNunc).
a. 225-cm2 tissue culture flasks.
b. 75-cm2 tissue culture flasks.
c. 25-cm2 tissue culture flasks.
d. 96-well tissue culture plates.
e. 40-µm sterile cell strainer.
24. 293 Cells (American Type Culture Collection [ATCC] cat. no. CRL 1573).
150 Gouze et al.
25. 293 Cre8 cells (generous gift of Dr. Stephen Hardy; see ref. 17).
26. Ψ5 Adenovirus (generous gift of Dr. Stephen Hardy; see ref. 17).
27. DOC lysis buffer: 100 mM Tris-HCl, pH 9.0, 20% (v/v) ethanol, 0.4% (w/v)
sodium deoxycholate.
28. 0.5 M, Spermine-HCl.
29. RNase A (10 mg/mL ; Sigma-Aldrich).
30. Sodium dodecyl sulfate (SDS), 10% (w/v).
31. 0.5 M EDTA, pH 8.0.
32. 2 M CaCl2.
33. Cesium chloride solutions:
a. 1.4 g/mL CsCl in 10 mM Tris-HCl, pH 7.9.
b. 1.2 g/mL CsCl in 10 mM Tris-HCl, pH 7.9.
34. Thin-walled ultraspeed centrifuge tubes (e.g., 10-mL polyallomer tubes; Sorvall).
35. Dialysis buffer: 10 mM Tris-HCl, pH 7.5, 200 mM NaCl, 1 mM EDTA, 4% (w/v)
sucrose.
36. Dialysis tubing, molecular-weight cutoff (MWCO) 50,000 (e.g., Spectra/Por,
Spectrum Laboratories).
37. Griess reaction.
a. Solution I: 1% (w/v) sulfanilamide, 2.5% (v/v) H3PO4.
b. Solution II: 0.1% (w/v) dihydrochloride N-ethylendiamine.
c. NaNO2, for standard curve.
d. Microplate reader, set at 550-nm wavelength.
3. Methods
The methods described below outline (1) the isolation of a specific cDNA
and its insertion into an adenoviral shuttle plasmid, (2) the generation and
amplification of recombinant adenovirus, (3) the isolation and culture of
articular chondrocytes and synoviocytes, (4) the adenoviral transduction of
these cells, and (5) measurement of cellular response to the transgene product.
The adenoviral vector system discussed in the following sections was devel-
oped by Hardy et al. (17) and makes use of Cre-mediated recombination among
an adenoviral shuttle vector, pAdlox, and an adenoviral backbone, Ψ5. The
resulting vector is a replication-deficient, type 5 adenovirus lacking the E1 and
E3 loci. The gene of interest is contained within a cytomegalovirus (CMV)-
driven expression cassette located in the site of the E1 domain.
To illustrate the practical application of these methods, the cDNA for human
GFAT will be used as a specific example.
3.1. Insertion of a Novel cDNA Into the Adenoviral Shuttle Plasmid
The shuttle plasmid described here, pAdlox (17), can serve two purposes.
The first is as a eukaryotic expression plasmid for verification of expression of
the gene product of the cDNA of interest. The second is as a shuttle plasmid for
Fig. 1. Schematic of pAdlox expression plasmid (generous gift of Dr. Stephen

Hardy; see ref. 19).
the generation of recombinant adenovirus (17). pAdlox contains the left-hand

inverted terminal repeat of the adenovirus, viral packaging signal (Ψ), a cDNA
expression cassette driven by the CMV promoter/enhancer, and finally, a loxP
recognition site for the Cre-recombinase (Fig. 1). Described below is (1) the
isolation of the cDNA for the human form of GFAT, (2) its insertion into
pAdlox, and (3) its propagation in a competent strain of E. coli, DH5α.
3.1.1. cDNA Cloning
1. Plate one 25-cm2 flask with cells that provide robust expression of the gene of
interest. For isolation of human GFAT, approx 105–293 cells were cultured in
DMEM/10% FBS/1% penicillin–streptomycin.
2. At confluence, remove the media and add 1 mL TRIzol reagent to cells; harvest
the lysate using a cell scraper (see Note 1).
3. Mix with 200 µL chloroform and centrifuge for 10 min at 13,000g, 4°C.
152 Gouze et al.
4. Remove the aqueous phase (top layer), and precipitate the RNA by addition of
2 vol of isopropanol.
5. Centrifuge for 10 min at 13,000g, 4°C, remove the supernatant, and resuspend
the RNA in 20 µL TE buffer.
6. Quantitate RNA by spectrophotometry, and freeze at –80°C.
7. For cDNA synthesis, heat 3 µg of total RNA to 65°C for 5 min and then place it
on ice. Reverse transcribe the RNA for 2 h at 37°C in a final volume of 20 µL of
reverse transcription buffer, plus 500 µM each of dATP, dCTP, dGTP, and dTTP,
10 mM dithiothreitol, 100 pmol random primer, 40 U Rnase OUT, and 200 U
MMLV RNase H– reverse transcriptase.
8. Amplify the cDNA of interest by incubating 2 µL of the cDNA from step 7 in a
PCR reaction of 50 µL consisting of PCR buffer plus 0.2 mM of each dNTP,
20 pmol of the forward and reverse oligodeoxynucleotide primers, and 2 U of Vent
DNA polymerase. Incubate the reaction for 5 min at 95°C and then 35 cycles of
1 min at 95°C, 1 min at 60°C, and 2 min at 72°C. For human GFAT, the forward
primer spanned nucleotides 32–54; the reverse primer was complementary to nucle-
otides 2197–2227 relative to the published human GFAT sequence (Genebank ac-
cession no. M90516) (see Note 2).
9. Following resolution of the reaction products on a 0.8% agarose gel, purify the
DNA fragment corresponding to the proper size using a QIAquick gel extraction
kit following the manufacturer’s instructions.
10. Dilute an aliquot of the purified DNA fragment 100–500 times in TE buffer.
11. Using the same conditions as in step 8, amplify by PCR 1 µL of the purified
DNA using sequence-specific forward and reverse oligodeoxynucleotide prim-
ers engineered to contain endonuclease restriction sites for convenient insertion
into the multiple cloning site of pAdlox (20 cycles) (see Fig. 2 and Note 3).
12. Digest pAdlox and the amplified DNA with the appropriate endonuclease restric-
tion enzymes (2 h, enzyme-specific temperature) (see Note 4). Gel-purify the frag-
ments as described in step 9.
13. Directionally ligate the DNA fragment and pAdlox (T4 ligase, 14°C overnight)
by standard methods (18).
14. Freeze sample at –20°C (optional).
3.1.2. Bacterial Transformation and Plasmid Selection

1. Transform DH5α cells with plasmid Adlox.GFAT using standard methods (18).
2. Plate bacterial cells on an LB agar dish containing ampicillin (50 µg/mL) and
incubate overnight at 37°C.
3. Carefully select single colonies and inoculate individual cultures of 10 mL LB
media with ampicillin (50 µg/mL). Incubate overnight at 37°C.
4. Following standard miniprep isolation of the plasmid DNA (18), screen each
isolate for the presence and orientation of the cDNA insert by strategic restriction
digestion.
5. Following identification of a suitable construct, prepare a large-scale culture
(500 mL) of the bacteria and incubate overnight at 37°C with vigorous shaking.
Fig. 2. Schematic of mutagenesis by PCR reaction. (A) cDNA cloning using 100%
homologous oligonucleotide. (B) Amplification using new oligonucleotides with a
BamHI or EcoRI site in their sequence. (C) Digestion with EcoRI and BamHI.
6. Purify the plasmid DNA from the bacteria using a Qiagen Tip 2500 Plasmid Kit
following the manufacturer’s instructions.
7. Freeze at –20°C (optional).
3.2. Recombinant Adenovirus

To generate recombinant virus for large-scale infections, replication-defi-
cient adenovirus is typically amplified in 293 cells, which supply in trans the
E1 gene products necessary for viral replication. Caution: stringent safety pro-
cedures should be followed while working with recombinant adenovirus (see
Note 5) (19). It is also strongly advised that the investigator be familiar with
the appropriate federal, state, and institutional regulations concerning micro-
biological safety.
154 Gouze et al.
Fig. 3. Construction of an E1-substituted adenovirus by using Ψ5 and a shuttle

vector (19). The gene of interest is directionally inserted into the pAdlox plasmid.
Cre-mediated recombination between the loxP site in the shuttle vector and the loxP
site in the Ψ5 adenoviral backbone not only allows generation of adenovirus carrying
the gene of interest but also avoids the propagation of non-recombinant viruses. The
packaging site is labeled Ψ. CMV, cytomegalovirus; ITR, internal tandem repeat.
The procedures described below outline (1) the preparation and amplification
of Ψ5 adenoviral DNA, (2) the generation of an adenoviral vector using a sys-
tem of Cre-lox recombination (17) (see Note 6), and (3) the large-scale prepara-
tion of recombinant adenovirus. For this, the pAdlox shuttle plasmid is
cotransfected with genomic DNA from the Ψ5 adenoviral backbone into 293
Cre8 cells. This cell line constitutively expresses Cre recombinase, which medi-
ates recombination between loxP sites in the shuttle plasmid and the Ψ5 aden-
oviral backbone, generating a novel recombinant virus. Negative pressure on
Ψ5 propagation in these cells is achieved by Cre-mediated intramolecular recom-
bination, which removes the Ψ packaging site from the Ψ5 backbone, prevent-
ing its insertion into the viral capsid. Thus, viral particles arising from the
transfection typically contain a very high percentage of recombinants (Fig. 3).
3.2.1. Preparation and Amplification of Ψ5 Adenoviral DNA

The 293 cell line, which does not express the Cre recombinase, is permis-
sive for replication of the Ψ5 adenovirus. Infection of these cells will allow
amplification of the virus for isolation of the genomic DNA and for the genera-
tion of Ψ5 viral stocks. Both procedures are described below.
1. Plate approx 5 × 105 293 cells in a 75-cm2 flask containing 15 mL DMEM/10%
FBS/1% penicillin–streptomycin.
2. At confluence, remove media and replace with a minimal volume (4 mL) of fresh
DMEM without serum, followed by 50 µL of Ψ5 viral lysate (see step 5). Incu-
bate for 2–4 h at 37°C, and then supplement with 10 mL of medium.
3. Monitor cells daily for signs of infection (see Note 7).
4. When the cells are rounded and begin to detach, harvest the cells and media using
a cell scraper.
5. Pellet the cells by centrifugation for 10 min at 2000g, 4°C and resuspend in 400 µL
of TE buffer. (Alternatively, the cell pellet can be resuspended in 5 mL of fresh
media and placed through three cycles of freeze-thaw [see Subheading 3.2.3.]).
This lysate is then frozen and stored at –80°C and serves as a stock for future
amplification of Ψ5 virus.)
6. For DNA purification, add 400 µL of DOC lysis buffer to the cell pellet. Mix by
pipeting 10–15 times through a 1-mL pipet and transfer the lysate to a microcen-
trifuge tube and add 8 µL of 0.5 M spermine-HCl. Incubate on ice for 10 min.
7. Centrifuge for 5 min at 13,000g to pellet cellular debris and genomic DNA.
8. Transfer the supernatant to a new microcentrifuge tube. Add 4 µL of 10 mg/mL
RNase A and incubate at 37°C for 10 min.
9. To release the viral DNA from the particles, add 60 µL of 10% SDS, 20 µL of
0.5 M EDTA, and 40 µL of 50 mg/mL pronase. Incubate for 60 min at 40°C.
10. Extract viral DNA with phenol/chloroform/isoamyl alcohol and then precipitate
using standard methods (18,19).
11. Wash the DNA pellet with 70% ethanol. Resuspend the DNA pellet in 25 µL TE
buffer. The viral DNA is now suitable for restriction digestion analysis or trans-
fection as needed. For example, digestion of 4–6 µL of the Ψ5 DNA with BsaBI
and resolution of the fragments in an 0.8% agarose gel generates a discreet restric-
tion pattern with a limited number of bands (see Note 8 and Fig. 4).
3.2.2. Generation of Recombinant Adenovirus

This procedure describes the cotransfection of the pAdlox shuttle plasmid
and the Ψ5 adenoviral backbone into 293 Cre8 cells, which results in the gen-
eration of a novel recombinant vector.
1. Plate approx 106 Cre8 cells in a 75-cm2 flask to 60% confluence in DMEM/10%
FBS/1% penicillin–streptomycin.
2. Cotransfect the Cre8 cells with 3 µg of pAdlox and 3 µg of purified Ψ5 adenovi-
ral DNA using standard methods (18,19) (see Note 9).
156 Gouze et al.
Fig. 4. Restriction mapping of Ψ5 DNA. Viral DNA was prepared, digested with
BsaBI, and then resolved by agarose gel electrophoresis. BsaBI cuts after the loxP
site in Ψ5, producing a 2.2-kb fragment that will be replaced by another band with a
recombinant DNA.
3. When prominent cytopathic effects are observed throughout the transfected cells
(~8–10 d), harvest the cells and media from the flask using a cell scraper. Pellet
by centrifugation 10 min at 2000g, 4°C and resuspend in 5 mL of fresh medium
in a 50-mL capped tube.
4. Freeze and thaw the harvested cell/media mixture three times to lyse the cells and
release the recombinant adenovirus. Store the lysate at –80°C.
5. To amplify, purify, or verify (see Subheading 3.2.1. and Note 8) the adenoviral
stock, thaw the viral lysate and mix 0.5 mL with 4 mL of fresh DMEM. Infect
new Cre8 cultures by replacing the medium with the medium/lysate mixture.
Return cells to the incubator for 4 h.
6. After the incubation, supplement the culture with fresh medium to normal cul-
ture conditions. Culture the cells until prominent cytopathic effects are observed
throughout the monolayer (~1–2 d). Harvest the cells and media, as in step 3, and
store the lysate at –80°C (see Note 10).
3.2.3. Large-Scale Preparation of Recombinant Adenovirus

1. Grow 293 cells to confluence in DMEM/10% FBS/1% penicillin–streptomycin
in two 225-cm2 tissue culture flasks.
2. Aspirate media and wash cells twice in prewarmed GBSS. Dilute adenoviral stock
in a small volume of GBSS and add to cells. Typically 5 × 107 plaque-forming units
(PFU) of virus are added to 5 mL of GBSS for each 75-cm2 flask (see Note 11).
3. Incubate at 37°C in a 5% CO2 incubator for 3–4 h. After incubation, remove the
viral solution and add 15 mL of DMEM/10%FBS/1% penicillin–streptomycin
per flask.
4. When prominent cytopathic effects are observed throughout the monolayer, har-
vest cells and media from flasks using a cell scraper and transfer into 50-mL
capped centrifuge tubes (see Subheading 3.2.1.4. and Note 7).
5. Centrifuge 10 min at 2000g, 4°C. Resuspend the cell pellets in 5–10 mL of GBSS
and then freeze-thaw the cells three times.
6. Add RQ1 DNase (80 U) to the freeze-thaw lysate and incubate at 37°C for 30 min.
7. Centrifuge lysate for 10 min at 2000g, 4°C, and aspirate supernatant.
8. Prepare a CsCl step gradient in prechilled polyallomer tubes containing 1/3 vol
of 1.4 g/mL CsCl (bottom layer), 1.2 g/mL CsCl (middle layer), and viral cell
lysate (top layer)
9. Centrifuge the tubes in a swinging bucket rotor 1h at 40,000g, 4°C.
10. Harvest the viral band from the tube in a minimal volume using a 16-gage needle
and 3 mL syringe (see Note 12). Dilute the harvested band at least twofold in
GBSS for recentrifugation.
11. Repeat steps 9–11 two more times.
12. Transfer the collected adenovirus fraction to dialysis bags and dialyze against
500 mL of prechilled dialysis buffer for at least 6 h at 4°C. Repeat one more time.
13. Following titration (see Note 11), aliquot the recombinant adenovirus into mul-
tiple, sterile Eppendorf tubes, to avoid decreasing viral titer by multiple freeze-
thawing of viral stocks. Freeze at –80°C.
3.3. Adenoviral Transduction of Chondrocytes and Synovial Fibroblasts

Described below are the steps used for (1) isolation and culture of
synoviocytes and chondrocytes from bovine tissues and (2) the adenoviral
transduction of these cultures with Ad.GFAT. Because of the quantity of tissue
that can be obtained, bovine tissue is a convenient source of cells. The harvest
of joint tissues from 1–2-wk-old calves and subsequent isolation of primary
cells within 4 h of slaughter typically results in sufficient recovery such that
experiments can be performed on primary cell isolates.
3.3.1. Isolation and Culture of Joint Cells
3.3.1.1. ARTICULAR CHONDROCYTES
1. From the bovine knee joint, shave cartilage pieces from the femoropatellar groove
and condyles using aseptic techniques. Mince the shavings using a razor blade or
scalpel.
2. Transfer the cartilage pieces into a 50-mL Falcon tube containing 10-mL sterile
filtered 0.15 M NaCl with 2% w/v pronase. Incubate for 1 h with gentle agitation
at 37°C.
3. Centrifuge digestion mixture for 5 min at 2000g, discard the supernatant, and
wash pelleted cartilage fragments twice with GBSS.
158 Gouze et al.
4. Resuspend the cartilage fragments in 30 mL Ham’s F-12 medium/10% FBS/1%

penicillin–streptomycin supplemented with collagenase (2% w/v) and incubate
at 37°C overnight with gentle agitation.
5. Filter the digestion mixture using a sterile 40-µm cell strainer (Falcon) to remove
the debris from the cell suspension. Collect the liquid suspension and pellet the
chondrocytes by centrifugation for 15 min at 2000g.
6. Plate the cells in a 25-cm2 flask in 3 mL Ham’s F-12/10% FBS/1% penicillin–
streptomycin. Wash the adherent cells 24 h later to remove debris, and return to
culture with fresh media.
3.3.1.2. SYNOVIOCYTES
1. Carefully harvest synovial membrane from the joint capsule under aseptic con-
ditions.
2. Using a razor blade, finely mince the recovered synovial tissue.
3. Transfer the minced tissue into 25 mL of Ham’s F12 containing 1.5 mg/mL of
collagenase and dispase and incubate for 1 h, at 37°C under gentle agitation.
4. Filter the digestion mixture through a 40-µm sterile cell strainer.
5. Collect the liquid suspension and centrifuge for 15 min at 2000g to pellet the
synovial cells.
6. Wash the cells twice in PBS and plate them in a 25-cm2 flask in 3 mL Ham’s F-12
medium/10% FBS/1% penicillin–streptomycin (see Note 13).
3.3.2. Adenoviral Transduction

Depending on the number of cells used in each experiment, achieving a high
percentage (>85%) of transduced synovial fibroblasts or chondrocytes may
require the use of large volumes of recombinant adenovirus. Mixing the aden-
oviral suspension with CaCl2 prior to infection has been shown to increase
the transduction efficiency of various cell types (20) (see Note 14). We have
recently confirmed that this is also effective for chondrocytes and synovial
fibroblasts. Described below are the procedures for CaCl2-enhanced transduc-
tion of these cells.
1. Seed two 25-cm2 flasks with 4 × 106 bovine articular chondrocytes or synovial
fibroblasts in Ham’s F-12/10% FBS/1% penicillin–streptomycin. This should
result in a near-confluent monolayer.
2. On the day of infection, wash the cells and add 3 mL of fresh medium.
3. Add the appropriate amount of adenoviral stock solution to achieve the desired
MOI (the multiplicity of infection is defined as the number of infectious viral
particles per cell) to 990 µL of serum-free Ham’s F-12 and mix thoroughly.
4. Slowly add 10 µL of a 2 M solution of CaCl2 with constant agitation (final con-
centration 20 mM), and then briefly vortex.
5. Leave to stand for 30 min at room temperature with occasional vortexing.
6. Add the 1 mL of viral/calcium mix to the 25-cm2 plate and leave infection to
proceed overnight.
Fig. 5. Nitric oxide(NO) production after interleukin-1 (IL-1) stimulation of

glutamine:fructose-6-phosphhate amidotransferase (GFAT)+ chondrocytes. Chondro-
cyte cultures were infected with Ad.GFAT at the indicated multiplicity of infection
(MOI). After 24 h, cells were treated with 5 ng/mL of human IL-1β. NO production
assessed by measurement of nitrite in the medium was measured 24 h later. Results are
expressed in mM nitrite (mean ± SD of three different assays). #, p < 0.05 vs
untransduced control; *, p < 0.05 vs untransduced cells treated by IL-1β. Infection
with Ad.GFAT decreased NO production after IL-1 stimulation in a dose-dependent
manner.
7. Aspirate the viral solution and feed the cells with 5 mL of fresh Ham’s F-12
medium/10% FBS/1% penicillin–streptomycin.
8. Verify GFAT mRNA expression in one flask using RT-PCR procedure (see
Note 15), and proceed to experiment further on the second flask.
3.4. NO Measurement
To assess the chondroprotective effect of the Ad.GFAT transduction, the
nitrate concentration in the culture medium can be measured 24 h after IL-1α
challenge (5 ng/mL) (see Fig. 5).
1. Produce a standard curve using sodium nitrite (concentration ranging from 0 to
200 µM).
2. Add 100 µL of standard or sample medium to a 96-well plate.
3. To each standard and sample add 50 µL of solution I (sulfanilamide 1% [w/v];
H3PO4 2.5%), followed by 50 µL of solution II (dihydrochloride N-ethylendi-
amine 0.1% [w/v]).
4. Read the absorbance within 15 min at 550 nm on an MR5000 microplate reader.
160 Gouze et al.
4. Notes
1. Precautions should be taken while working with RNA to avoid RNase contami-
nation. It is strongly recommended to wear disposable gloves and work only with
DNase/RNase-free material and supplies.
2. When the coding sequence is not entirely available in the database, 3' and 5'
RACE-PCR methods can be used to complete the missing part of the sequence
(21) (Life Technologies).
3. To achieve RT-PCR amplification of the coding sequence of any gene, two oli-
gonucleotides have to be designed, allowing amplification of all sequences that
extend from the translation start site through the stop codon. A second PCR ampli-
fication using a new set of oligonucleotide primers containing convenient endo-
nuclease restriction sites allows a directional insertion within the polylinker or
multiple cloning sites of the plasmid. Alternatively, a blunt-ended ligation can be
performed. However, it will be necessary to confirm the orientation of insertion
by specific digestion of the plasmid. Following insertion into the plasmid, the
cDNA should be sequenced to verify its accuracy.
4. The choice of the endonuclease restriction enzymes depends on (1) their presence in
the polylinker site of pAdlox plasmid and (2) their absence in the cDNA sequence
of the gene of interest. Concerning GFAT cDNA, we choose EcoRI and BamHI.
5. Wild-type adenovirus has been associated with upper respiratory disease in humans,
and recombinant virus is highly infectious for numerous human tissues. First-gen-
eration adenoviral vectors are replication-defective and are only able to replicate in
permissive cells. The 293 cell lines used here contain the left-hand 11% of the wild-
type adenoviral genome, which contains the E1 locus. Because these proteins are
provided in trans, they can complement the E1 deletion in the adenoviral vector.
Recombinant adenovirus is a biosafety level 2 hazardous agent. Whenever possible,
work should be performed in a class II biological safety cabinet. In addition, proper
protective clothing and eyewear should be worn at all times. Solid waste should be
rinsed with 10% bleach solution and placed in biohazard containers. Liquid waste
should be decanted or aspirated into a large container of bleach. Likewise, surfaces
on which all virus work was performed should be decontaminated with 10% bleach.
Extra caution should be used during the handling of sharp objects.
6. Multiple methods can be utilized in the generation of recombinant adenovirus.
Several companies have now developed techniques to generate such recombi-
nant vectors (i.e., AdEasy System, Stratagene). The procedure described in this
chapter represents only one method among others.
7. Cytopathic effects appear as an increase of granularity, a rounding aspect, and a
tendency of the cells to detach from the plates. Monitoring the lytic infection
closely is key to obtaining a high yield of adenovirus.
8. Any appropriate restriction enzyme can be used. In the case of BsaBI, the 2249
band is from the left-hand end of the Ψ5 DNA and contains the expression cas-
sette. Thus, its size will be altered by the insertion of a specific cDNA in a recom-
binant virus. Usually by the second or third passage in CRE8 cells, the Ψ5 helper
virus is no longer detectable by restriction analysis of the DNA.
9. Primary digestion of pAdlox with SfiI is recommended. This will release the
adenoviral component of the plasmid, and in this linear form, recombination
with the Ψ5 backbone is enhanced several fold. This will generate two bands, a
2500-bp fragment and a fragment of 1600 bp plus the cDNA insert (19). How-
ever, if the cDNA of interest contains internal SfiI recognition sites, such as the
GFAT cDNA, this step can be omitted.
10. When generating new recombinant adenovirus, it is essential to determine that it
expresses a functional gene product. This can be done by enzyme-linked
immunosorbent assay, Western blot, or bioassay when available. If there is no
existing quantitative assay for the protein of interest, transgenic expression can
be measured at the RNA level by RT-PCR (see also Note 15).
11. To measure virus particle concentration, mix a 50-µL aliquot with 950 µL of
GBSS and determine A260 by spectrophotometry. (One A260 is approximately
equal to 1012 viral particles/mL.) The percentage of infectious virions typically
ranges between 1 and 10% of the total number of viral particles. The infectious
titer may be determined by performing a plaque assay on confluent cultures of
293 cells.
12. Typically two bands will be seen near the interface of the 1.2- and 1.4-g/mL CsCl
layers. The lower band containing the infectious particles is the band that is
collected.
13. To isolate the fibroblastic synoviocyte population preferentially, cells can be pas-
saged three times to eliminate nonadherent, type A, macrophage-like cells. The
type B synovial fibroblast can then be split into 75-cm2 flasks and grown until
80% confluence.
14. Compared with adenovirus alone, adenovirus-CaCl2 coprecipitates increase bind-
ing of virus to cells. Although the mechanism involved is not yet understood,
increased binding is probably responsible for the significant increase in transgene
expression (20).
15. Transcriptional expression of the human cDNA delivered by the adenovirus can
be distinguished from that of the endogenous homolog of the xenogenic host cell
using RT-PCR and a modification of the method of Khiri et al. (22). When avail-
able, a restriction site polymorphism between the coding sequences of the two
species can allow strategic digestion such that the DNA amplification products
of only one of the two species is digested into two fragments, whereas the other
species remains uncut.
References
1.
1 Pincus, T. and Callahan, L. F. (1992) Early mortality in RA predicted by poor
clinical status. Bull. Rheum. Dis. 41, 1–4.
2.
2 Felson, D. T., Lawrence, R. C., Hochberg, M. C., et al. (2000) Osteoarthritis: new
insights. Part 2: treatment approaches. Ann. Intern. Med. 133, 726–737.
3.
3 Smalley, W. E., Ray, W. A., Daugherty, J. R., and Griffin, M. R. (1995) Nonste-
roidal anti-inflammatory drugs and the incidence of hospitalizations for peptic
ulcer disease in elderly persons. Am. J. Epidemiol. 141, 539–545.
162 Gouze et al.
4.
4 Tamblyn, R., Berkson, L., Dauphinee, W. D., et al. (1997) Unnecessary prescrib-
ing of NSAIDs and the management of NSAID-related gastropathy in medical
practice. Ann. Intern. Med. 127, 429–438.
5.
5 Reginster, J. Y., Deroisy, R., Rovati, L. C., et al. (2001) Long-term effects of
glucosamine sulphate on osteoarthritis progression: a randomised, placebo-con-
trolled clinical trial. Lancet 357, 251–256.
6.
6 Muller-Fassbender, H., Bach, G. L., Haase, W., Rovati, L. C., and Setnikar, I.
(1994) Glucosamine sulfate compared to ibuprofen in osteoarthritis of the knee.
7.
7 Bassleer, C., Rovati, L., and Franchimont, P. (1998) Stimulation of proteoglycan
production by glucosamine sulfate in chondrocytes isolated from human osteoar-
thritic articular cartilage in vitro. Osteoarthritis Cartilage 6, 427–434.
8.
8 Gouze, J. N., Bordji, K., Gulberti, S., et al. (2001) Interleukin-1β down-regulates
the expression of glucuronosyltransferase I, a key enzyme priming glycosami-
noglycan biosynthesis: influence of glucosamine on interleukin-1β-mediated
effects in rat chondrocytes. Arthritis Rheum. 44, 351–360.
9.
9 Gouze, J. N., Bianchi, A., Becuwe, P., et al. (2002) Glucosamine modulates IL-1-
induced activation of rat chondrocytes at a receptor level, and by inhibiting the
NF-κ B pathway. FEBS Lett. 510, 166–170.
10.
10 Sandy, J. D., Gamett, D., Thompson, V., and Verscharen, C. (1998) Chondrocyte-
mediated catabolism of aggrecan: aggrecanase-dependent cleavage induced by inter-
leukin-1 or retinoic acid can be inhibited by glucosamine. Biochem. J. 335, 59–66.
11.
11 Shikhman, A. R., Kuhn, K., Alaaeddine, N., and Lotz, M. (2001) N-acetylglu-
cosamine prevents IL-1 β-mediated activation of human chondrocytes. J. Immunol.
166, 5155–5160.
12. Chang, Q., Su, K., Baker, J. R., Yang, X., Paterson, A. J., and Kudlow, J. E. (2000)
Phosphorylation of human glutamine:fructose-6-phosphate amidotransferase by
cAMP-dependent protein kinase at serine 205 blocks the enzyme activity. J. Biol.
Chem. 275, 21,981–21,987.
13. Gouze, J., Ghivizzani, S., Gouze, E., et al. (2004) Adenovirus-mediated gene trans-
fer of glutamine/fructose-6-phosphate amidotransferase antagonizes the effects of
interleukin 1 beta in rat chondrocytes. Osteoarthritic Cartilage 12, 217–224.
14.
14 Ghivizzani, S. C., Lechman, E. R., Kang, R., et al. (1998) Direct adenovirus-
mediated gene transfer of interleukin 1 and tumor necrosis factor alpha soluble
receptors to rabbit knees with experimental arthritis has local and distal anti-
arthritic effects. Proc. Natl. Acad. Sci. USA 95, 4613–4618.
15 Ghivizzani, S. C., Oligino, T. J., Glorioso, J. C., Robbins, P. D., and Evans, C. H.
15.
(2001) Direct gene delivery strategies for the treatment of rheumatoid arthritis.
Drug. Discov. Today 6, 259–267.
16.
16 Yao, Q., Glorioso, J. C., Evans, C. H., et al. (2000) Adenoviral mediated delivery
of FAS ligand to arthritic joints causes extensive apoptosis in the synovial lining.
J. Gene Med. 2, 210–219.
17 Hardy, S., Kitamura, M., Harris-Stansil, T., Dai, Y., and Phipps, M. L. (1997)
17.
Construction of adenovirus vectors through Cre-lox recombination. J. Virol. 71,
1842–1849.
18. Sambrook, J., Fritsch, E. and Maniatis, T. (eds.) (1989) Molecular Cloning: A Labo-
ratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
19.
19 Palmer, G. D., Gouze, E., Gouze, J. N., Betz, O. B., Evans, C. H., and Ghivizzani,
S. C. (2003) Gene transfer to articular chondrocytes with recombinant adenovi-
rus. Methods Mol. Biol. 215, 235–246.
20.
20 Fasbender, A., Lee, J. H., Walters, R. W., Moninger, T. O., Zabner, J., and
Welsh, M. J. (1998) Incorporation of adenovirus in calcium phosphate precipi-
tates enhances gene transfer to airway epithelia in vitro and in vivo. J. Clin.
Invest. 102, 184–193.
21.
21 Frohman, M. A., Dush, M. K., and Martin, G. R. (1988) Rapid production of full-
length cDNAs from rare transcripts: amplification using a single gene-specific
oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85, 8998–9002.
22. Khiri, H., Reynier, P., Peyrol, N., Lerique, B., Torresani, J., and Planells, R. (1996)
Quantitative multistandard RT-PCR assay using interspecies polymorphism. Mol.
Cell Probes 10, 201–211.
Chondrocyte Metabolism Changes 165
13
Changes of Chondrocyte Metabolism In Vitro
An Approach by Proteomic Analysis
Anne-Marie Freyria and Michel Becchi
Summary
Changes in chondrocyte metabolism in vitro using different support systems and under dif-
ferent culture conditions were studied with a proteomic approach. Qualitative and quantitative
modifications in the synthesis of chondrocyte proteins were investigated using two-dimen-
sional (2D) gel electrophoresis. This technique provided a simple way to visualize the most
abundant chondrocyte proteins. Proteins were identified after in-gel proteolysis with trypsin
and matrix-assisted laser desorption ionization-time of flight mass spectrometry, using peptide
mass fingerprinting. Tryptic peptide masses were measured and matched against a computer-
generated list from the simulated trypsin proteolysis of a protein database (SwissProt).
Key Words: Proteomics; cartilage; chondrocyte culture; 2D electrophoresis; in-gel protein

digestion; MALDI-TOF; peptide-mass fingerprinting.
1. Introduction
Articular cartilage is a tissue that has a limited capacity for repair after trauma
or diseases such as osteoarthritis (OA) (1). Among the numerous changes that
occur during the course of OA, a modulation of the chondrocyte phenotype that
can be reproduced in vitro is observed (2). Most in vitro studies have analyzed
chondrocyte growth and the synthesis of cartilage-specific matrix components
by chondrocytes of different origins grown under different culture conditions:
in monolayer, in a 3D environment and in explants (3,4). More recently, quali-
tative and quantitative modifications in the synthesis of chondrocyte proteins
have been investigated using 2D electrophoresis and microsequencing, to get
insights into the mechanisms involved in chondrocyte differentiation and carti-
lage regeneration (5–7). The applications of proteome analysis in a clinical con-
text were also reported in the secretion medium of OA cartilage explants (8) and
in the synovial fluids and plasma from OA patients (9).
165
166 Freyria and Becchi
Two-dimensional electrophoresis is a powerful technique for separating

proteins from a mixture according to two physical parameters: charge and
molecular weight. The identification of proteins may be carried out using
immunolabeling when antibodies are available, Edman microsequencing to
obtain amino acid sequences that will match known sequences in databases,
or mass spectrometry. Matrix-assisted laser desorption (MALDI) (10) and
electrospray ionization (ESI) (11) are two ionization methods frequently used
in protein mass spectrometry analyses (12,13). Total protein mass is insuffi-
ciently discriminating to identify proteins with confidence; rather, proteins
should first be converted into shorter peptides by proteolysis. The usual
method involves excision of gel spots of interest, in-gel digestion using trypsin
(the enzyme that cleaves the C-terminal to the lysyl and arginyl residues), and
finally, mass spectrometric analysis of the peptides produced (14–16). There
is a direct relationship between mass spectrometric data and peptide amino
acid sequences, and at this stage, two methods are available for protein iden-
tification:
1. Peptide mass fingerprinting (PMF), which compares an experimental peptide
mass profile with a theoretical profile calculated from the known sequences in a
protein database (17,18).
2. Peptide fragmentation information generated by tandem mass spectrometry (MS/
MS). Mass spectra produced by collision-induced decomposition of selected pep-
tide ions are compared with theoretical tandem mass spectra in a database (19).
We have used PMF with a MALDI-TOF (time-of-flight) instrument. The
efficiency of a PMF search strongly depends on the accuracy of peptide mass
measurements. This technique may be sufficient to characterize proteins from
completely sequenced genomes, as the masses of all theoretical peptides can
be calculated precisely. The PMF strategy is generally completed by com-
bining it with MS/MS to generate additional sequence information. MALDI-
TOF mass spectrometry can offer limited information via postsource decay
(PSD) mass spectra (20,21), but, for small proteins and for proteins that gen-
erate a limited number of peptides in the classical working mass range (600–
4000) from trypsin proteolysis, PSD can help confirm their identification.
2. Materials
2.1. Chondrocyte Culture Equipment
1. Hemacytometer (Malassez).
2. Centrifuge (up to 3800g).
3. Inverted-phase contrast microscope.
4. 12.5-cm2 Sterile culture flasks.
5. Orbital shaker (Heidolph, Schwabach, Germany).
2.2. Chondrocyte Culture

When not indicated, prepare the solutions from analytical grade reagents
and dissolve in deionized water.
1. Phosphate-buffered saline (PBS; Sigma, Saint Quentin Fallavier, France): one
pouch dissolved in 1 L of distilled water yields 0.01 M sodium phosphate,
0.138 M NaCl, 0.0027 M KCl, pH 7.4.
2. 0.4% Trypan Blue solution (Sigma).
3. Cartilage: metacarpophalangeal joints of calves less than 6 mo old, obtained from
a local slaughterhouse less than 3 h after death.
4. Chondrocyte isolation solution: 0.2% (w/v) hyaluronidase type I-S (Sigma) dis-
solved in PBS.
5. 0.1% (w/v) Collagenase type IA (Sigma) + 0.1% (w/v) dispase II (Roche Applied
Bioscience, Meylan, France) dissolved in PBS (see Note 1).
6. Cell culture medium: NCTC 109 medium + RPMI-1640 medium (Sigma; 50/50,
v/v) containing 0.25% glutamine (Sigma), 10% fetal calf serum (Seromed,
Strasbourg, France), 1% penicillin–streptomycin (PS), and 0.4% amphotericin B
(stock solutions; Gibco BRL Life Technologies).
7. Biomaterials: bovine type I collagen sponges (Coletica, Lyon, France) cross-
linked using n-hydroxysuccinimide/1-ethyl-3(-3-dimethylaminopropyl)carbodi-
imide hydrochloride (NHS/EDC) methods.
2.3. Protein Preparation and Separation Equipment

1. Sterile 15-mL tubes, Eppendorf tubes, and Petri dishes (35 × 10 mm).
2. Reswelling tray (Amersham Biosciences) for the rehydration of the immobiline
drystrips (see Subheading 2.4.)
3. First-dimension separation: MultiphorTM II system and EPS 3500 XL power sup-
ply (Amersham Biosciences). Temperature control provided by a Haake F3 ther-
mostatic circulator (Haake, Bioblock Scientific).
4. Second-dimension separation: power supply (1000/500) and a Criterion cell (Bio-
Rad, Life Science).
2.4. Protein Preparation and Separation Reagents

1. Immobiline drystrip, pH 4.0–7.0, 11 cm (Bio-Rad), immobilized pH gradient
(IPG) buffer pH 4.0–7.0 (Amersham Biosciences).
2. Criterion precast gel 10% acrylamide (Bio-Rad).
3. Collagenase in PBS (1.5 mg/mL).
4. Trypsin/EDTA (Gibco-BRL Life Technologies) diluted with PBS to obtain final
concentrations of 0.1%/0.04%.
5. Dithiothreitol (DTT) and iodoacetamide (Sigma).
6. Molecular mass marker proteins: sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) standard (Bio-Rad).
7. Protein lysis buffer: 8 M urea (Bio-Rad), 1 M thiourea (Sigma), 4% CHAPS (3-
[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate; Sigma), 40 mM
Tris base. This buffer can be stored in 250-µL aliquots at –20°C for up to 4 mo.
Do not heat above 30°C to dissolve the different components.
8. Drystrip rehydration buffer: 8 M urea, 2% CHAPS, 0.2% IPG buffer (same pH
range as the drystrip), 50 mM DTT (Sigma), and a trace of bromophenol blue
(Bio-Rad). This buffer can be stored in 250-µL aliquots at –20°C for up to 4 mo.
9. SDS equilibration buffer: 0.375 M Tris-HCl, pH 8.8, 6 M urea, 20% glycerol, 2%
SDS (Sigma), and a trace of bromophenol blue. The buffer can be stored in 12-mL
aliquots at –20°C for up to 4 mo.
10. SDS equilibration buffer containing a reducing agent (equilibration buffer 1):
dissolve 130 mM DTT in 6 mL equilibration buffer. Prepare the solution just
prior to use.
11. SDS equilibration buffer containing an alkylating agent (equilibration buffer 2):
dissolve 135 mM iodoacetamide in 6 mL equilibration buffer. Prepare the solu-
tion just prior to use.
12. Running buffer for the second dimension gel: 0.025 M Tris, 0.192 M glycine, 1%
SDS, pH 8.3. This buffer can be stored at 4°C for up to 6 mo.
13. Coomassie blue solution: 0.25 g of Coomassie R250 (Roth, Lauterbourg, France)
dissolved in 400 mL methanol, 100 mL acetic acid, and 500 mL deionized water.
14. Destaining solution: 30% methanol, 10% acetic acid in deionized water.
2.5. In-Gel Enzymatic Digestion Equipment

1. Heating block for microcentrifuge tubes with shaker (type: Thermomixer com-
fort, Eppendorf, Fisher Scientific).
2. Water bath at 37°C.
3. Vacuum concentrator (e.g., Speed-Vac Savant, Avantec).
4. Centrifuge (5000–7000g; e.g., Micro 7, Fisher Scientific).
5. Laboratory rotating mixer (Vortex, Fisher Scientific).
6. Scalpel.
7. Tweezers.
8. Glass sheet, 15 × 15 cm.
9. Polypropylene microcentrifuge tubes (0.5 and 1.5 mL).
10. Ultrafine tips (gel loader type pipet tips, VWR International, Strasbourg, France).
2.6. In-Gel Enzymatic Digestion Reagents

When not indicated prepare the solutions from analytical grade reagents.
1. Milli-Q grade water (Millipore).
2. Ethanol/water (20/80, v/v).
3. 100 mM Ammonium bicarbonate (NH4HCO3), pH 8.0, stock solution, aliquoted
in 1.5-mL Eppendorf tubes and kept at –20°C. Working solutions: 20 and 50 mM
NH4HCO3 in water.
4. Sequencing-grade modified trypsin (20 µg/vial; Promega).
5. Acetonitrile.
6. Formic acid/water (5/95, v/v).
7. Acetonitrile/water (50/50, v/v).
2.7. MALDI–MS Equipment

1. MALDI-TOF mass spectrometer equipped with delayed source extraction, oper-
ating in linear and/or reflector mode. For PSD capability for MS/MS experiments,
we use a Voyager-DE™ PRO MALDI-TOF from Applied Biosystems.
2. ZipTipSCX (Millipore) pipet tips containing strong cation exchange resin.
2.8. MALDI–MS Reagents

1. Trifluoroacetic acid (TFA; Sigma); stock solution: 1% TFA in water (v/v). Work-
ing solution: 0.1% in water (v/v).
2. Matrix: α-cyano-4-hydroxycinnamic acid (HCCA, purified; Laser Bio Labs,
Sophia-Antipolis, France): a saturated solution at 5 mg/0.5 mL of acetonitrile/
water (50/50 v/v) containing 0.1% TFA is prepared every d.
3. Methanol.
4. Ammonium hydroxide, 25% stock solution in water.
5. Elution solution for ZipTipSCX: 5% ammonium hydroxide/30% methanol in Milli-
Q-grade water (in a 1.5-mL centrifuge tube: 500 µL H2O, 300 µL CH3OH, and
200 µL 25% NH4OH).
3. Methods
3.1. Protein Sample Preparation From Cell Cultures
3.1.1. Cell Collection and Solubilization From Monolayer Culture
1. Isolate chondrocytes from minced cartilage by enzymatic digestion with a mix-
ture of 0.2% hyaluronidase and 0.2% collagenase/dispase in PBS at 37°C for 4 h
with gentle magnetic stirring.
2. Wash isolated cells three times in culture medium and count them with an hema-
cytometer. Check viability using the trypan blue solution (1 vol cell suspension +
0.1 vol trypan blue solution).
3. Seed the cells at high-density culture (0.6 × 106 cells/cm2) and grow them in an
incubator at 37°C in a closed flask with 5 mL of culture medium for up to 1 mo,
changing the medium every 3 d.
4. Harvest cells (at the chosen time of analysis):
a. Wash twice in PBS and incubate for 2 min at 37°C in 0.1% trypsin/EDTA
(100 µL/flask).
b. Detach cells with a cell scraper, add 2 mL of culture medium, and suspend the
cells by repeated pipeting in a 15 mL tube, so as to obtain a single-cell sus-
pension.
c. Add 10 mL of culture medium to the suspension and count the cells.
d. Centrifuge for 6 min at 1400g, discard the supernatant, resuspend the pellet in
1 mL PBS and place it in an Eppendorf tube. Centrifuge and re-suspend the
pellet in lysis buffer at 50 µL per 4 million cells (see Subheading 2.4.7.).
3.1.2. Cell Collection and Solubilization From a 3D Culture

1. Seed collagen sponges with 1 × 106 or 10 × 106 cells per sponge (2.3 × 106 or 23
× 106 cells per cm3). Culture them in a 37°C incubator in 5 mL of culture medium
per sponge, in a closed flask, changing the medium every 3 d.
2. Harvest the cells:
a. Cut each sponge into four pieces with a scalpel blade in a Petri dish.
b. Place the pieces in a tube with 1 mL 0.2% hyaluronidase in PBS. Shake for
15 min on the orbital shaker (30 rpm) in the incubator at 37°C.
3. Collect the liquid phase in a 15-mL tube and centrifuge for 4 min at 1400g. Dis-
card the supernatant, add 1 mL of PBS to the pellet, and resuspend the cells. Keep
this tube (T1) for the cell count.
4. Treat the sponge phase with 1 mL of collagenase at 1.5 mg/mL in PBS for 15 min
at 37°C with gentle shaking. Collect the fluid phase, centrifuge, and resuspend the
pellet in 2 mL PBS. Keep the tube (T2) for the cell count. Treat the remaining
sponge phase in the same way (T3). Wash the sponge phase with 5 mL of culture
medium for 15 min at 37°C and collect the liquid phase (T4). Pool all four tubes
into a 15-mL tube, and count the cells with a hemacytometer. Centrifuge the tube
for 4 min at 1400g, and discard the supernatant. Resuspend the pellet in 1 mL
PBS, and then transfer it to an Eppendorf tube and centrifuge. Resuspend the pel-
let in cell lysis buffer (50 µL per 4 million cells) and store at –20°C for the assay.
3.2. 2D Electrophoresis (see Note 2)
3.2.1. First Dimension: Isoelectric Focusing (IEF)
1. Rehydrate the drystrip gel, pH 4.0–7.0, with a mixture of the sample in lysis
buffer (50 µL) and rehydration buffer (230 µL), place in one groove of the
reswelling tray, and protect by a layer of paraffin oil (3 mL per drystrip) for 12 h
at room temperature. (The number of gel strips required corresponds to the
samples to be compared in one experiment.)
2. Set the temperature of the Haake F3 thermostatic circulator to 19°C prior to the
start of IEF.
3. Place the rehydrated drystrips into the tray of the Multiphor II system and verify
that the anodic terminus of the drystrip faces the anode side of the Multiphor II
system. Place the electrodes at both ends of the strips. Pour about 80 mL of par-
affin oil into the tray. Start the IEF according to the parameters given in Table 1,
using the gradient mode of the EPS 3500 XL power supply.
4. When IEF is complete, discard the paraffin oil and remove the electrodes and the
focused gel strips from the tray using clean forceps.
5. Store the focused gel strips in clean individual tubes (15 mL) at –20°C until the
second dimension run.
3.2.2. Drystrip Gel Equilibration

1. Take the tube containing the focused gel out of the freezer.
2. Add equilibration buffer 1 and place the tube on a shaker for 15 min at room
temperature. Remove equilibration buffer 1 from the tube.
Table 1
Running Conditions for Drystrips, pH 4.0–7.0, 11 cm
Time Voltage Current (mA) Power (W)
30 min 0–300 1 5
30 min 300 1 5
90 min 300–3500 1 5
5 ha 3500 1 5
aAdjust the time of the last step of the program to obtain 20.6 kVh at
the end.
3. Add equilibration buffer 2 and place the tube on the shaker for another 15 min at
room temperature.
4. Rinse the gel briefly with distilled water to remove excess equilibration buffer
and place it on a piece of filter paper on the edge of one long side for 2–3 min.
(Bending the gel helps to keep it on the long side.)
3.2.3. Second Dimension

1. Prepare the Criterion precast gel for the second dimension according to the
manufacturer’s instructions while the focused gel strip is equilibrating.
2. Rinse the top of the gel with running buffer and leave buffer in the two wells
(one small one for the molecular weight marker and one long one for the
focused gel).
3. Load the focused gel strip on top of the Criterion gel, with the plastic sheet facing
the top glass plate. Push down on the gel with a spatula, taking care to avoid
creating or trapping air bubbles where the two gels meet. Add 10 µL of molecular
mass marker protein to the small well.
4. Insert the gel cassette into the electrophoresis chamber of the Criterion cell con-
taining 800 mL of fresh running buffer. Fill the upper chamber with 60 mL of
running buffer.
5. Run the machine for 10 min at 100 V and for 45–60 min at 200 V (ensure that the
bromophenol blue migrates to the end of the SDS-PAGE gel; see Note 3).
3.2.4. Staining of the Gel

1. Immediately after electrophoresis, rinse the gel with distilled water for 1–2 min
with gentle shaking to remove SDS.
2. Soak the gel in 200 mL of a fresh solution of Coomassie blue for 20–30 min.
Rinse briefly with water, and then soak in 200 mL destaining solution until the
desired contrast is achieved. Keep gel at 4°C in water for a few d until in-gel
protein digestion (see Fig. 1 and Note 4).
Fig. 1. Two-dimensional electrophoretic separation of intracellular proteins extracted

from bovine chondrocytes grown in 3D culture for 3 wk (initial seeding 10 × 106 cells).
The protein sample was treated following the procedures given in Subheadings 3.1.2.
for protein extraction and 3.2. for protein separation. This figure gives an example of
the effectiveness of direct extraction from collagen sponges. The pattern compares with
a published pattern obtained after protein extraction from a monolayer culture (6).
For spots (1) and (2) see the MALDI-TOF mass fingerprinting results in Figs. 2 and 3,
respectively.
3.3. In-Gel Protein Digestion (see Note 5)

3.3.1. Excision of Protein Spots
1. Clean work areas and all utensils (scalpel, tweezers, glass sheets, and so on) with
methanol.
2. Identify the spot of interest on the gel and excise it with a scalpel. Cut it as close
as possible to the protein spot (see Note 6).
3. Cut the excised spot into roughly 1-mm3 pieces and transfer them to a clean,
labeled 0.5-mL microcentrifuge tube.
3.3.2. Destaining of Coomassie Blue-Stained Gel Pieces

1. Add 400 µL 20 mM NH4HCO3 to gel pieces. Let the tubes stand for 15 min at
30°C (in Thermomixer without shaking or water bath). Remove liquid using gel
loading pipet tips.
2. Add 400 µL of CH3CN/H2O (50/50, v/v) and incubate for 15 min at 30°C.
Remove solvent using gel loading pipet tips. Repeat this step until the blue
coloration disappears.
3. Add 400 µL H2O and incubate for 15 min at 30°C. Remove water using gel load-
ing pipet tips and dry the gel pieces in a vacuum centrifuge. If a large volume of
gel is used, shrink it by adding CH3CN before drying.
4. Keep dried gel pieces at –20°C until the next step (see Note 7).
3.3.3. In-Gel Digestion

1. Stock solution of trypsin: dissolve 20 µg of trypsin (see Subheading 2.6.) in
200 µL 50 mM NH4HCO3 at 25–30°C for 10 min (see Note 8).
2. Distribute stock solution into 0.5 mL polypropylene centrifuge tubes in 10–20
µL aliquots and keep at –20°C. Thaw the required trypsin solution at room
temperature before use. The final concentration of trypsin should be 0.1 µg/µL.
3. Apply trypsin solution directly onto gel pieces. The trypsin/protein ratio in the
gel should be about 1:10. Use 0.5–4 µL of trypsin solution, depending on the
intensity of the protein spot (i.e., protein concentration): from 0.5 µL for small
and lightly colored spots and up to 4 µL for large and intense spots.
4. Centrifuge for 2 min at 5000–7000g to allow trypsin to penetrate inside the gel
pieces.
5. Complete rehydration with enough 50 mM NH4HCO3 buffer to just cover the
gel pieces. At this stage avoid the absence or excess of NH4HCO3 buffer (see
Note 9).
6. Digest for 6 h at 37°C. Check buffer volume during digestion (especially after
about 1 h of digestion) and add fresh buffer, if necessary, to keep the gel pieces
completely covered and wet.
3.3.4. Extraction of Tryptic Peptides

1. Spin and shake the microcentrifuge tubes in the Thermomixer Comfort
(Eppendorf) at 30°C for 8 min. Centrifuge for 2 min at 4600–5000g (8–9000 rpm).
2. Remove supernatant (digest buffer) with a gel loader pipet tip and keep it in a
new labeled 0.5-mL microcentrifuge tube.
3. Extract successively with CH3CN, HCOOH/H2O (5/95, v/v) and CH3CN/H2O (50/
50, v/v). For each extraction step, add approximately the same volume of solvent as
the volume of trypsin solution plus buffer used for digestion. After adding solvent,
shake for 10 min at 30°C in the Thermomixer Comfort (Eppendorf) at 1300 rpm,
centrifuge, and remove solvent extract with a gel loader pipet tip. Pool all corre-
sponding extracts in the same labeled microcentrifuge tube.
4. Keep the extract solutions at –20°C until mass spectrometry analysis.
3.4. MALDI-TOF Sample Preparation

3.4.1. Loading of the MALDI-TOF Target
1. Matrix preparation: see Subheading 2.8.2.
2. Dry down extract solutions of peptides in a vacuum concentrator.
3. Redissolve peptides in 12–15 µL 0.1% TFA, vortex, and then centrifuge.
4. Add 1 µL of sample solution and 1 µL matrix solution to the target and leave to
dry in a gentle stream of air. (We use the fans of the MALDI-TOF instrument.)
3.4.2. Sample Clean-Up on ZipTipSCX

1. Equilibrate the ZipTipSCX pipet tip by drawing up 10 µL 0.1% TFA (use a 10-µL
pipet; see Note 10). Expel the TFA solution. Repeat this process three times.
2. Bind the peptides from the remaining sample solution used in Subheading 3.4.1.
by carrying out 12–15 aspirate-dispense cycles of the entire sample (see Note 11).
3. Wash the ZipTipSCX by drawing up 10 µL 0.1% TFA into the tip and expelling it.
Repeat this process five times.
4. Desorb bound peptides by aspirating 2.5 µL elution solution (see Subheading 2.8.).
Deposit eluates directly onto the MALDI-TOF target and pipet up and down three
times.
5. Dry spots in a gentle air stream.
6. Overlay each peptide spot with 1 µL matrix solution (see Subheading 2.8.).
3.5. MALDI-TOF–MS Analysis

MALDI-TOF mass spectra are acquired in the 700-5000-Dalton mass range
using a minimum of 200 shots of laser per spectrum.
3.5.1. Internal Calibration
Delayed ion extraction and reflectron MALDI-TOF mass spectrometer
equipment allow sufficient resolution to consider monoisotopic masses of sin-
gly charged ions [M+H]+. The accuracy of mass measurements depends on
the calibration procedures. External calibration is performed with a mixture
of known peptides covering the mass range applied (as close as possible to
but well separated from the sample deposit). We use internal calibration with
autolysis ion fragments of trypsin at m/z 842.5100, 1045.5642, 2211.1046,
and 2283.1807, which allow a peptide mass tolerance of 30 ppm for peptide
mass measurements.
3.5.2. Peptide Mass Fingerprinting
1. Follow the manufacturer’s instructions for mass spectra acquisitions (see Sub-
heading 2.7., Note 12, and Fig. 2). MALDI produces only singly charged [M+H]+
peptide ions and is tolerant toward complex mixtures, permitting the mass of
many different peptides to be recorded.
2. Carry out a first assay on the sample extract solution (see Subheading 3.4.1.).
Fig. 2. (A) MALDI-TOF mass spectrum of a tryptic digest of a 2D gel separated

protein. #, matched peptide ions [M + H]+ for the ATP synthase β-chain (SwissProt
accession number P00829). T, trypsin autolysis peptide fragments. (See following page
for Fig. 2B.)
3. If the mass spectrum shows a very low ion count then carry out a second assay
after ZipTipSCX treatment (see Subheading 3.4.2.).
4. Obtain a list of monoisotopic [M+H]+ masses of tryptic peptides for each pro-
tein spot. Remove trypsin autolysis peptides from this list (i.e., peptides used for
internal calibration; see Subheading 3.5.1.) and contaminants from the gel (see
Note 13).
3.5.3. Peptide-Mass Searches

1. Several search engines available on the internet, and these can all be found on the
proteomic server of the Swiss Institute of Bioinformatics at http://www.expasy.
org/: MS-Fit (ProteinProspector), Mascot, PeptideSearch, Pepident, and so on.
Fig. 2. (Continued from previous page) (B) MS-Fit search results for peptide
mass fingerprinting for the first proposition.
We have MS-Fit in-house and we also use Mascot on the web to confirm results.
Whichever one you use, there are several parameters to specify:
a. Database: we use the SwissProt database for a PMF search. The NCBInr data-
base may be used to find other sequence homologies or relevant propositions.
b. Species: select Mammals from the suggested species.
c. MW of protein: input a 30 kDa range from the estimated MW from the 2D
gel (see Note 14).
d. Protein pI: select all because proteins may be post-translationally modified
(phosphorylation, glycosylation, and so on).
e. Digest: trypsin, with 1 as the maximum number of missed cleavages.
f. Cysteine modification: input carbamidomethylation, as the protein is reduced
and alkylated by iodoacetamide (see Subheading 3.2.2.).
g. Possible modifications: we generally choose oxidation of methionine and
acetylation of the protein N-terminus.
h. Mass spectrometry parameters: MALDI-TOF instrument, peptide monoiso-
topic masses with a tolerance of 30 ppm.
i. Minimum number of peptides required to match: 4 is a good starting point.
2. Under reported hits, don’t expect all the masses that you put into the search engine
to be tryptic peptides from the protein spot, as some of them will be contaminants.
It is not only the number of hits, i.e., the number of matched peptides, that is
important. There is also a MOWSE score that uses larger fragments, which are
statistically more significant when measuring probability. The protein recovery
percentage from the matched peptides is also a useful parameter.
3.5.4. Postsource Decay MALDI–MS

When the confidence of a PMF search is too low (Fig. 3), additional
sequence information on the observed peptides can be provided from PSD
spectra (Fig. 4). Our MALDI-TOF instrument is fitted with a variable voltage
reflector, which involves a long and fastidious recording of the spectra (as the
reflector has to be scanned at different voltage values), as well as the different
spectra being added and then reconstructed by the software. PSD has other
drawbacks with regard to ESI–MS/MS techniques: ion selection is not very
specific and is made over a relatively large mass range 20 Dalton, and there
is no way to control or improve the fragmentation. We only consider very
abundant parent ions [M + H]+ for PSD analysis.
PSD MALDI–MS spectra are generally very complicated, as all the classi-
cal peptide ion fragments containing both N- and C-termini are observed (for
nomenclature, see ref. 23) as well as internal fragments. We use instrument
software (a protein fragmentation calculator) to interpret PSD spectra. Starting
with the supposed sequence, the software calculates all possible fragmenta-
tions, and then the theoretical ion fragments are compared with those observed
on the PSD MALDI–MS spectrum (see Note 15).
4. Notes
1. Prepare the appropriate volume just before cell extraction or cartilage digestion,
as enzyme solutions are less effective after the freeze-thaw step.
Fig. 3. (A) MALDI-TOF mass spectrum of a tryptic digest of a 2D gel separated

protein. (B) (opposite page) The MS-Fit search results show only five matched pep-
tides (labeled # on the mass spectrum) for the 27-kDa heat shock protein (SwissProt
accession number P42929) from Canis familiaris (CANFA). The abundant ions at m/z
1413.74 and 1819.93 are not identified. Protein identification is completed using a post
source decay mass spectrum (see Fig. 4).
2. As the goal of the separating steps is to be able to identify the proteins present in
the different samples using mass spectrometry analysis, it is recommended to
wear gloves (powder-free or nitrile) at each stage, first to eliminate contamina-
tion with keratin and second, to be able to use silver staining methods safely (see
ref. 22 for details) when a higher sensitivity than Coomassie blue is desired.
3. Make sure to run the gels from corresponding experiments in a similar way so as
to be able to compare gel patterns by image analysis, if necessary.
4. When the purpose of the study is to compare different chondrocyte culture condi-
tions, scan the wet gels (e.g., with the Personal Densitometer SI and Image Quant
software from Amersham Biosciences) in order to make significant comparisons
and qualitative and quantitative analyses of the proteins of interest.
5. One of the main problems encountered is keratin contamination of samples. You

must wear a lab coat and gloves to avoid any contact with gel electrophoresis
samples, reagents, and in-gel digestion equipment by skin, fingers, or any keratin
source (woolen sweater, and so).
6. When excising the spot, try to include only the protein-containing gel by cutting
close to the edge of the spot. Excise another piece of roughly the same size in a
gel region without protein spots (to be used as a blank control).
7. In cases of silver-stained gels, destaining is carried out as described in ref. 24.
8. Make sure that the lyophilized trypsin powder is at the bottom of the vial, as it
sometimes sticks under the vial cap.
Fig. 4. Post source decay (PSD) mass spectrum of the peptide at m/z 1163.62 from
the tryptic digest shown in Fig. 3. The immonium ions (P, L, Q, F, and R), C-terminal
ions y2, y4, and y7, N-terminal ions b2, b3, b4, and b5 and internal fragment ions
confirm the sequence LFDQAFGLPR. This peptide belongs to the 27-kDa heat shock
protein family (see MS-Fit results in Fig. 3). As the bovine form of this protein has
probably not been fully sequenced, the peptides at m/z 1413.74 and 1819.93 in Fig. 3A
might be attributed to as yet uncharacterized sequences.
9. Lack of buffer leads to the gel pieces drying if left above the liquid surface, and
an excess of buffer means a dilution of trypsin outside the gel. In both cases this
could lead to the incomplete digestion of in-gel proteins.
10. Information and technical services concerning ZipTip SCX are available at
www.millipore.com/ziptip.
11. Carry out each movement slowly, to give enough contact time between the sample
and the ZipTipSCX packing. Avoid passing air through the tip.
12. General information on MALDI-TOF–MS can be found on the following Web-
site: http://www.srsmaldi.com.
13. Gel contaminants can easily be detected by comparison with a blank spot gel (see
Note 6). The corresponding ions appear with a mass default on the first decimal:
m/z 855.03, 871.03, 1060.05, and so on.
14. Do not be too accurate in the initial estimation of protein MW because, if it is
slightly out, you may exclude the protein from the search parameters. MW of a
protein parameter in Mascot is not really a cutoff value, as in MS-Fit, for search-
ing in the SwissProt database. In this case, the input value must not be greater
than the sum of the peptide masses listed.
15. It is also possible to match ion fragments from the PSD MALDI–MS spectrum to
MS/MS databases using a customized search engine such as MS-Tag
(ProteinProspector).
References
1.
1 Hunziker, E. B. (2002) Articular cartilage repair: basic science and clinical
progress. A review of the current status and prospects. Osteoarthritis Cartilage
10, 432–463.
2. Benya, P. D. and Brown, P. D. (1986) Modulation of the chondrocyte phenotype
in vitro, in Articular Cartilage Biochemistry (Kuettner, K. E., Schleyerbach, R.,
and Hascall, V. C., eds.), Raven, New York, NY, pp. 219–233.
3. Ronzière, M. C., Farjanel, J., Freyria, A. M., Hartmann, D. J., and Herbage, D.
(1997) Analysis of types I, II, III, IX and XI collagens synthesized by fetal bovine
chondrocytes in high-density culture. Osteoarthritis Cartilage 5, 205–214.
4.
4 Roche, S., Ronzière, M. C., Herbage, D., and Freyria, A. M. (2001) Native and
DPPA cross-linked collagen sponges seeded with fetal bovine epiphyseal
chondrocytes used for cartilage tissue engineering. Biomaterials 22, 9–18.
5.
5 Freyria, A. M., Ronziere, M. C., Boutillon, M. M., and Herbage, D. (1995) Two-
dimensional electrophoresis of intracellular and secreted protein synthesized by
fetal bovine chondrocytes in high-density culture. Electrophoresis 16, 1268–1272.
6.
6 Freyria, A. M., Ronziere, M. C., Boutillon, M. M., and Herbage, D. (1995) Effect
of retinoic acid on protein synthesis by foetal bovine chondrocytes in high-den-
sity culture: down-regulation of the glucose-regulated protein, GRP-78, and type
II collagen. Biochem. J. 305, 391–396.
7.
7 Freyria, A. M., Ronzière, M. C., Roche, S., Rousseau, C. F., and Herbage, D.
(1999) Regulation of growth, protein synthesis and maturation of foetal bovine
chondrocytes grown in high-density culture in the presence of ascorbic acid,
retinoic acid and dihydrocytochalasin B. J. Cell. Biochem. 76, 84–98.
8. Hermansson, M., Bolton, M., Alexander, S., and Wait, R. (2002) Application of
proteomics to analysis of chondrocyte gene expression in normal and osteoarthri-
tis cartilage, in XVIIIth Federation of European Connective Tissue Societies Meet-
ing Abstracts, Brighton, England, p. 46.
9.
9 Sinz, A., Bantscheff, M., Mikkat, S., et al. (2002) Mass spectrometric proteome
analyses of synovial fluids and plasmas from patients suffering from rheumatoid
arthritis and comparison to reactive arthritis or osteoarthritis. Electrophoresis 23,
3445–3456.
10.
10 Karas, M. and Hillenkamp, F. (1988) Laser desorption ionisation of proteins with
molecular masses exeeding 10,000 daltons. Anal. Chem. 60, 2299–3201.
11.
11 Fenn, J. B., Mann, M., Meng, C. K., Wong, S. F., and Whitehouse, C. M. (1989)
Electrospray ionisation for mass spectrometry of large biomolecules. Science 246,
64–71.
12.
12 Yates, J. R. III (1998) Mass spectrometry and the age of the proteome. J. Mass
Spectrom. 33, 1–19.
13.
13 Mann, M., Hendrickson, R. C., and Pandey, A. (2001) Analysis of proteins and
proteomes by mass spectrometry. Ann. Rev. Biochem. 70, 437–473.
14.
14 Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric
sequencing of proteins from silver-stained polyacrylamide gel. Anal. Chem. 68,
850–858.
15.
15 Jungblut, P. and Thiede, B. (1997) Protein identification from 2-D gels by MALDI
mass spectrometry. Mass Spectrom. Rev. 16, 145–162.
16. Jensen, O. N., Wilm, M., Shevchenko, A., and Mann, M. (1999) Sample prepara-
tion methods for mass spectrometric peptide mapping directly from 2-DE gels, in
Methods in Molecular Biology: 2-D Proteome Analysis Protocols (Link, A. J.,
ed.), Humana, Totowa, NJ, pp. 513–530.
17.
17 Henzel, W. J., Billeci, T. M., Stults, J. T., Wong, S. C., Grimley, C., and Watanabe,
C. (1993) Identifying proteins from two-dimensional gels by molecular mass
searching of peptide fragments in protein sequence databases. Proc. Natl. Acad.
Sci. USA 90, 5011–5015.
18. Beavis, R. C. and Fenyö, D. (2000) Data base searching with mass spectrometric
information. Trends Biochem. Sci. Suppl. Proteomics: A Trends Guide, pp. 22–27.
19. Kinter, M., and Sherman, N. E. (eds.) (2001) Protein Sequencing And Identifica-
tion Using Tandem Mass Spectrometry. John Wiley & Sons, New York, NY.
20.
20 Spengler, B. (1997) Post-source decay analysis in matrix-assisted laser desorption/
ionization mass spectrometry of biomolecules. J. Mass Spectrom. 32, 1019–1036.
21. Courschesne, P. L. and Patterson, S. D. (1999) Identification of proteins by ma-
trix-assisted laser desorption/ionization mass spectrometry using peptide and frag-
ment ion masses, in Methods in Molecular Biology: 2-D Proteome Analysis
Protocols (Link, A. J., ed.), Humana, Totowa, NJ, pp. 487–511.
22. Rabilloud, T. (1999) Silver staining of 2-D electrophoresis gels, in Methods in
Molecular Biology: 2-D Proteome Analysis Protocols (Link, A. J., ed.), Humana,
Totowa, NJ, pp. 297–305.
23.
23 Biemann, K. (1990) Appendix 5. Nomenclature for peptide fragment ions (posi-
tive ions). Methods Enzymol. 193, 886–887.
24. Gharahdaghi, F., Weinberg, C. R., Meagher, D. A., Imai, B. S., and Mische, S. M.
(1999) Mass spectrometric identification of proteins from silver-stained polyacry-
lamide gel: a method for the removal of silver ions to enhance sensitivity. Elec-
trophoresis 20, 601–605.
Chondrocyte Analysis by Flow Cytometry 183
14
Analysis of Chondrocyte Functional Markers
and Pericellular Matrix Components by Flow Cytometry
Gust Verbruggen, Jun Wang, Lai Wang, Dirk Elewaut, and Eric M. Veys
Summary
Flow cytometry has been used as a procedure to characterize the phenotype and function of
human articular cartilage cells cultured as monolayers or in gelled artificial matrices. Proce-
dures allowing intact cells with their cell-associated matrix, to be obtained have been described.
Appropriate monoclonal antibodies have allowed plasma membrane-associated proteins, e.g.,
growth factors and cytokine receptors, as well as the cell-associated extracellular matrix mac-
romolecules, to be studied. Intracellular compounds have been traced in permeabilized cells
after blocking of their intracellular transport and secretion mechanisms. We report the use of
fluorescent dye-labeled monoclonal antibodies or specific binding proteins against extracellu-
lar matrix compounds such as hyaluronan, aggrecan, types I and II collagen, and fibronectin.
The autocrine and paracrine growth factor and cytokine pathways considered include the insu-
lin-like growth factor-1 (IGF-1)/IGF receptor I (IGFRI), and the transforming growth factor-β1
(TGF-β1)/TGF-β receptor II (TGF-βRII) cascades, as well as the interleukin-1α/β (IL-1α/β)/
interleukin-1 receptors I and II (IL-1RI and II) systems. Catabolic enzymes that mediate extra-
cellular matrix turnover, e.g., some matrix metalloproteinases and their natural inhibitors, were
also studied. Finally, flow cytometry was used to assess the results of some pharmacological
interventions on the aforementioned variables in cultured chondrocytes.
Key Words: Flow cytometry; chondrocyte culture; agarose; alginate; monoclonal antibod-
ies; IGF; IGFRI; IL-1; IL-1RI; IL-1RII; TGF; TGFRII; MMP; TIMP; hyaluronan; aggrecan;
type I collagen; type II collagen; fibronectin.
1. Introduction
Flow cytometry is used as a standard procedure to characterize the pheno-
type of circulating lymphomyeloid cells. The deleterious effect of any pro-
teolytic digestion procedure on plasma membrane antigens limits application
of this method to cells isolated from tissues or cultures. When articular
chondrocytes are considered, the availability of fluorescent agents has enabled
183
184 Verbruggen et al.
researchers to use flow cytometry mainly to study cell viability (1), prolifera-
tion and cell cycle characteristics (2,3), respiratory activities, and free radical
production (4). The availability of specific monoclonal antibodies (MAbs) and
the development of nonproteolytic procedures of cell release has allowed flow
cytometry to be used for study of some plasma membrane-associated antigens,
e.g., cytokine receptors (5,6), cellular adhesion molecules (7,8), HLA classes I
and II antigens (9), membrane-bound peptidases (10), and intracellular
cytokines (11). Flow cytometry has also been used to study the synthesis of
extracellular matrix (ECM) molecules (12).
Articular cartilage is composed of a hydrated extensive ECM in which
chondrocytes are embedded. The intercellular matrix of cartilage is composed
of two compartments: the cell-associated matrix (CAM) lying close to the chon-
drocyte and, adjacent to this, the interterritorial matrix (ITM) (13). The CAM
is a constant part of the ECM (14,15) in which the macromolecular compounds
are metabolized or turned over in a particular way (16). Newly synthesized
aggrecans have been shown to reside in the CAM for short periods with a higher
rate of aggrecan turnover than in the ITM (17). The neosynthesized CAM mac-
romolecules leave the territorial matrix in a later stage, to diffuse to the ITM
(16). The ITM forms the largest domain of the intercellular matrix. One of the
advantages of chondrocyte culture in alginate or agarose is the reversibility of
the gelled condition of these matrices, allowing the study of the different inter-
cellular compartments surrounding the chondrocyte in vitro. Our studies on the
homeostasis of the ECM of articular cartilage have been conducted on
chondrocytes that maintained their original phenotype in vitro when cultured
in 3D culture systems, e.g., in gelled alginate or agarose.
In 3D culture, isolated chondrocytes resynthesize their CAM and ECM
within 1–2 wk (18). The CAM is characterized by a high content of large
proteoglycan aggregates, bound to the cell via the interaction of hyaluronan
with CD44-like receptors (19). Chondrocyte cell membranes also exhibit sev-
eral proteins with binding affinity for collagen, including anchorin CII (20),
heparan sulfate proteoglycan (20,21), and chondronectin (22). This enables
the use of flow cytometry to study the expression of the ECM macromol-
ecules belonging to the pericellular domain of the chondrocytes. The pres-
ence or absence of chondrocyte-specific aggrecan and collagens could then
be used as phenotypic markers of cartilage cells maintained in different in
vitro conditions.
Other possible targets in flow cytometry studies are the autocrine and
paracrine growth factor and cytokine pathways that control ECM metabolism,
which have been successfully studied by flow cytometric methods. The recep-
tor systems have been assessed on intact cells, whereas the cytokine or growth
factor levels have been assayed inside permeabilized cells (18,23).
This chapter presents procedures for the isolation and culture of human
chondrocytes as a monolayer or in gelled matrices, along with approaches for
flow cytometric analysis of a series of markers, pericellular components, and
effector molecules. Also presented is an overview of the investigations per-
formed on chondrocytes by flow cytometric analysis.
2. Materials
Chondrocytes are cultured in an incubator set at 37°C in water-saturated air
at 5% CO2.
2.1. Chondrocyte Isolation
1. Dulbecco’s modified Eagle’s medium (DMEM) with Na-pyruvate and low (1 g/L)
glucose (Gibco).
2. Stock solution of 104 U/mL penicillin, 10 mg/mL streptomycin and 0.025 mg/
mL fungizone (PSF; Gibco).
3. Fetal bovine serum (FBS; Gibco)
4. Solution of 0.25% hyaluronidase (from sheep testes; Sigma) in DMEM with 1%
PSF. Filter-sterilize through a 0.22-µm membrane.
5. Solution of 0.25% pronase E (from Streptomyces griseus; Sigma) in DMEM with
1% PSF. Filter-sterilize through a 0.22-µm membrane.
6. Solution of 0.25% collagenase (from Clostridium histolyticum; Sigma) in DMEM
with 1% PSF and 10% FBS. Filter-sterilize through a 0.22-µm membrane.
2.2. Chondrocyte Culture

2.2.1. Culture in Monolayer
1. DMEM with 1% PSF and 10% FBS.
2. Tissue culture flasks (25-cm2 surface area; Nunc) or 50-mm-diameter culture
dishes (6-well culture plates; Nunc).
2.2.2. Culture in Agarose Gel

1. Ultralow-gelling-temperature agarose, type IX, gelling at <15°C and remelting
at >50°C (Sigma). Prepare a solution of 3% agarose in distilled water, and then
autoclave it twice at 120°C for 15 min. This solution can be stored at 4–8°C
before use.
2. Nunc cryotubes (3.8-mL capacity). Coat each cryotube with 100 µL of melted
3% agarose, and then allow to gel at 4–8°C for 15 min.
3. Double concentrated (2X) DMEM: one package of DMEM in powder (Gibco),
normally used for 1 L of medium, is solubilized in 450 mL of distilled water. Add
3.7 g of NaHCO3, check the pH, bring to 500 mL, and filter-sterilize through a
0.22-µm membrane. Add ascorbic acid (Sigma) and FBS at twice the desired
final concentration.
4. Nutrient medium: DMEM plus 1% PSF, 10% FBS, and 50 µg/mL ascorbic acid.
2.2.3. Culture in Alginate Gel

1. 2X Hanks’ balanced salts solution (HBSS): one package of HBSS in powder
without calcium and magnesium (Sigma), normally used for preparation of 1 L,
is solubilized in 450 mL of distilled water. Add 0.35 g of NaHCO3, check the pH,
bring to 500 mL and filter-sterilize through a 0.22-µm membrane.
2. Alginate, low viscosity, from Macrocystis pyrifera (Sigma). Prepare a 4% solu-
tion in distilled water, and then sterilize by autoclaving at 120°C for 15 min. This
solution can be stored at 4°C and brought to 37°C before use.
3. Calcium chloride solution, 102 mM in distilled water. Filter sterilize through a
0.22-µm membrane.
4. Isotonic saline solution: 0.15 M sodium chloride in distilled water. Filter sterilize
through a 0.22-µm membrane.
5. Nutrient medium: DMEM plus 1% PSF, 10% FBS, and 50 µg/mL ascorbic acid.
6. 6-Well plates.
2.3. Chondrocyte Collection for Flow Cytometry
1. Monensin (GolgiStop™, BD Biosciences Pharmingen).
2. Cytofix/Cytoperm Plus™ Kit (BD Biosciences Pharmingen).
2.3.1. Chondrocytes in Monolayers
1. EDTA solution: 1 mmol/L EDTA in HBSS without Ca2+/Mg2+.
2. Cell scraper.
2.3.2. Harvesting From Agarose Gel
1. Agarase solution: 50 U/mL of agarase (agarose 3-glycanohydrolase from Pseudo-
monas atlantica; Sigma) in phosphate buffered saline (PBS). Prepare fresh on
the day of use.
2.3.3. Harvesting From Alginate Gel
1. Sodium citrate solution: 55 mM tri-sodium citrate dihydrate, pH 6.8, 0.15 M NaCl.
2.4. Flow Cytometry
1. Antibodies: all the antibodies are conjugated with fluorescein isothiocyanate
(FITC). They can be prepared as previously described (24), or by using commer-
cial kits such as the FluoroTag™ FITC conjugation kit (Sigma Aldrich).
a. The optical density of the FITC-conjugated antibodies is measured at 495 nm
(FITC) and 280 nm (protein) by spectrophotometry.
b. The degree of conjugation is calculated according to the following equation:
antibody/FITC = (2.87 × A495)/(A280 − 0.35 × A495). Optimal conjugation
requires a ratio between three and five.
c. Antibodies can be alternatively conjugated with phycoerythrin (PE) as previ-
ously described (25), or by using the Phycolink® phycoerythrin conjugation
kits by Prozyme.
d. The conjugated monoclonal antibodies (MAbs) are used in a direct immunof-
luorescent staining protocol for flow cytometry. Appropriate FITC- or PE-
labeled isotype-matched mouse or rabbit IgG1 or IgG 2α can be used as a
negative control.
e. The different monoclonal and polyclonal antibodies, or specific reporter pro-

teins used thus far in our flow cytometry protocols are given in Table 1 (see
also refs. 26–28).
2. Propidium iodide (PI): solution of 5 µg/mL in PBS.
3. PBS containing 0.1% sodium azide and 0.2% bovine serum albumin (BSA).
4. Quantum Simply Cellular Microbead Kit (Sigma).
5. Cell fluorimeter: FACSort, (Becton Dickinson) with CELLQuest software, or
equivalent apparatus.
3. Methods
3.1. Isolation of Chondrocytes
Human articular chondrocytes are isolated as described elsewhere (29,30),
with a few modifications. For studies on healthy cartilage cells, only donors
who have died after a short illness are selected. Cartilage samples from people
who have received corticosteroids or cytostatic drugs are not used.
1. Sample visually intact cartilage from the femur condyles, and dice it in small
(1-mm3) fragments.
2. Liberate the cells from their extracellular matrix through sequential enzymatic
digestion at 37°C. First add to the tissue the solution of 0.25% hyaluronidase in
DMEM with 1% PSF for 120 min.
3. Replace the hyaluronidase solution with 0.25% pronase in DMEM with 1% PSF
and incubate for 90 min.
4. Wash the tissue twice with DMEM with 1% PSF and 10% FBS and incubate
overnight in the same medium (see Note 1).
5. Aspirate the medium, and then solubilize the tissue matrix by incubating for 3–6 h
in 0.25% collagenase in DMEM 1% PSF plus 10% FBS (see Note 2).
6. Centrifuge the cells at 500g for 10 min and wash the pellet with DMEM 1% PSF
plus 10% FBS. Repeat this steps two more times. Count the viable cells on a
hemocytometer by the trypan blue exclusion test. Usually, 150 × 10 6
chondrocytes can be obtained from the femoral condyles of one adult individual
and more than 95% of the cells are viable after isolation.
3.2. Culture of Chondrocytes

Cells can be cultured in monolayers or in gelled artificial matrices.
3.2.1. Culture in Monolayer
To prepare chondrocytes in monolayer for flow cytometry, seed 5 × 104
cells in 0.5 mL of DMEM with 1% PSF and 10% FBS as nutrient medium per
1 cm2 of culture dish. Allow the cells to attach for 48 h (see Note 3). Medium
is replaced every 2–3 d. Chondrocytes are seeded in low-density cultures and
phenotypic analysis by flow cytometry must be done when cells are still
subconfluent, usually after 4 d. This culture procedure makes possible to obtain
single cells embedded in their CAM.
188
Table 1
Antibodies or Protein-Specific Reporter Molecules a
MAb clone Isotype Source

Aggrecan AD11-2A9 IgG1 Mouse Biosource Europe, Nivelles, Belgium
Type I collagen I-8H5 IgG2a Mouse ICN Biochemicals, Aurora, OH
Type II collagen II-4C11 IgG1 mouse ICN Biochemicals
188
Fibronectin polyclonal IgG Rabbit Chemicon International, Harrow, UK

IGFRI 33255.111 IgG1 Mouse R&D Systems, Abingdon, UK
IL-1RI 35730.111 IgG1 Mouse R&D Systems
IL-1RII 34141.11 IgG1 Mouse R&D Systems
TGF-βRII polyclonal IgG Rabbit Santa Cruz Biotechnology,Santa Cruz, CA
IGF-1 AHG0014 IgG1 Mouse Biosource Europe, Nivelles, Belgium
IL-1α 624B3F2 IgG1 Mouse Biosource Europe
IL-1β 8516.311 IgG1 Mouse R&D Systems
TGF-β1 9016.2 IgG Rabbit R&D Systems
Verbruggen et al.
MMP-1 36665.111 IgG Rabbit R&D Systems
MMP-3 55-2A4 IgG1 Mouse Oncogene Research Products, Boston, MA
Chondrocyte Analysis by Flow Cytometry
MMP-13 181-15A12 IgG1 Mouse Oncogene Research Products
TIMP-1 7-6C1 IgG1 Mouse Oncogene Research Products
TIMP-3 136-13H4 IgG1 Mouse Oncogene Research Products
Hyaluronan bHABP b
IgG1 negative control IgG1 Mouse R&D Systems
Becton Dickinson, San Jose, CA
IgG negative control IgG Rabbit Santa Cruz Biotechnology
IgG2a negative control IgG2a Mouse R&D Systems
a The antihuman chondrocyte-specific aggrecan MAb was shown to react specifically with the G1-domain of the invariable
hyaluronan-binding region of the human aggrecan molecule. No crossreactivity with other known matrix components could be de-
tected (34,35). Antihuman type I and II collagen MAbs were raised in BALB/c mice after immunization with human placental type I
189
collagen and type II collagen from human costal cartilage, respectively. On Western blots antitype I collagen was shown to react
specifically with the a-chains and the triple helix molecular form of type I collagen, and the antitype II collagen MAb specifically
reacted with the a2(II) chain of type II collagen (36). Matrix metalloproteinase (MMP)-1 antibody was against both pro- and active
form of MMP-1. MMP-3 antibody recognized latent and active MMP-3 and reacted with MMP-3/tissue inhibitor of metalloproteinase
(TIMP)-1 complexes. MMP-13 MAb recognized both latent and active MMP-13. TIMP-1 antibody reacted with free TIMP-1 and
MMP/TIMP-1 complexes. Anti-TIMP-3 recognized both glycosylated and unglycosylated TIMP-3 (23). IGF, insulin-like growth
factor; IGFRI, IGF receptor I; IL, interleukin; IL-R, IL receptor; TGF, transforming growth factor.
b Biotinylated hyaluronic acid binding protein (bHABP) was a kind gift from Dr. J. Melrose (Raymond Purves Research Labora-
tories, University of Sidney, Australia) and was used to trace hyaluronan in the CAM. Binding of bHABP to hyaluronan in the CAM
was followed by a second-step avidin-FITC (Becton Dickinson) staining.
189
3.2.2. Culture in Agarose Gel

Chondrocytes are cultured in gelled agarose as previously described (31)
with some modifications (32,33). Chondrocyte suspension cultures are esta-
blished in 1.5% agarose in Nunc cryotubes (see Note 4).
1. Melt 3% (double concentrated) agarose gel and keep it at 37°C.
2. Mix equal volumes of melted 3% agarose and 2X DMEM supplemented with
20% FBS and 100 µg/mL of ascorbic acid.
3. Adjust the density of the chondrocyte suspension to 5 × 107 cells/mL in DMEM
supplemented with 1% PSF and 10% FBS.
4. Mix 10 vol of agarose in DMEM with 1 vol of the chondrocyte suspension, then
pipet 300-µL aliquots in coated cryotubes, and allow to gell at 4–8°C for 15 min.
The final cell density is around 1.5 × 106 chondrocytes per culture.
5. Add 3 mL of culture of nutrient medium (DMEM plus 1% PSF, 10% FBS, and 50
mg/mL ascorbic acid).
6. Incubate the cultures for 2 wk, replacing the nutrient medium three times a wk
(see Note 5).
3.2.3. Culture in Alginate Gel
Chondrocyte cultures in alginate beads are prepared as described elsewhere
(34) with some modifications.
1. Adjust the density of the chondrocyte suspension to 1 × 107 cells/mL in 2X HBSS
without calcium and magnesium.
2. Carefully mix 1 vol of chondrocyte suspension with an equal vol of 4% alginate.
Chondrocyte concentration is 5 × 106 cells/mL in 2% alginate.
3. Slowly drip the suspension through a 23-gage needle into a 102 mM calcium
chloride solution. The tip of the needle should be positioned about 1 cm above
the surface of the CaCl solution. Allow complete polymerization of the beads for
10 min at room temperature.
4. Aspirate the calcium chloride solution, and wash the beads three times with iso-
tonic saline solution.
5. Transfer the beads to a six-well plate in 4 mL/well of nutrient medium (DMEM
plus 1% PSF, 10% FBS, and 50 µg/mL of ascorbic acid). A culture of 20 beads
consists of about 1 × 106 chondrocytes. Replace the nutrient medium twice
weekly for 7–14 d (see Note 5).
3.3. Collection of Chondrocytes for Flow Cytometry

If the aim of the study is to evaluate the expression of proteins inside the
cells (cytokines, growth factors, matrix metalloproteinases (MMPs) and their
tissue inhibitors of metalloproteinases (TIMPs), and others, chondrocyte cul-
tures must be ended by incubation with 0.67 µL of monensin (GolgiStop)/mL
of medium for 5 h, in order to block protein transport from the Golgi apparatus
(18,23,35). No monensin treatment will be needed if the aim of the study is to
evaluate the expression of protein-epitopes located on the cell membrane or in
the CAM (e.g., receptors for growth factors and cytokines; integrins and other
receptors of CAM components; CAM macromolecules such as aggrecan, type
II collagen, and fibronectin).
3.3.1. Chondrocytes in Monolayer
Chondrocytes cultured on dishes are detached by EDTA treatment. The limi-
tation of EDTA treatment is the difficulty of dissociating confluent monolayer-
cultured chondrocytes embedded in a dense matrix. This procedure is therefore
applied to subconfluent cultures and makes it possible to obtain single
chondrocytes embedded in their CAM (35).
1. Aspirate the medium from the culture dish, replace with 20 mL of EDTA solu-
tion, and incubate at 37°C for 30 min.
2. If necessary, help cell detachment using a cell scraper, and collect the suspension
in a tube. Wash the dish with 10 mL of PBS, and add to the tube. For further
processing, see Subheading 3.4.
3.3.2. Chondrocytes in Agarose
Chondrocytes cultured in agarose are released by digestion of the gel with
agarase (36) (see Note 6).
1. Aspirate the medium from each cryotube, add 2 mL of agarase solution, and
incubate at 37°C for 1 h.
2. Collect the suspension in a tube. Wash the cryotube with 1 mL of PBS, and add
the wash to the tube. For further processing, see Subheading 3.4.
3.3.3. Chondrocytes in Alginate
Chondrocytes cultured in alginate are released by solubilization of the gel
by a Ca2+-chelating agent (see Note 6) (18).
1. Aspirate the medium from each culture well, wash the beads once with PBS with-
out Ca2+ and Mg2+, then add 3 mL of sodium citrate solution, and incubate for 10
min at 25°C.
2. Collect the suspension in a 10-mL tube. Wash the dish with 3 mL of PBS, and
add the wash to the tube. For further processing, see Subheading 3.4.
3.4. Preparation of Chondrocytes for Flow Cytometry

The chondrocytes harvested from the different culture systems are prepared
for further processing with antibodies. If the aim of the study is to evaluate the
expression of intracellular proteins, monensin-treated chondrocytes are fixed
and permeabilized using the Cytofix/Cytoperm Plus Kit, according to the
manufacturer’s instruction.
Cells (either in native form, or fixed and permeabilized) are then prepared
for incubation with MAbs (18,23,35), as follows:
1. Centrifuge the cells at 800–1500g for 10 min.
2. Wash the pellets with the chondrocytes and their CAM with PBS.
3. Resuspend about 2 × 105 chondrocytes in 100 µL of PBS containing 0.1% sodium
azide and 0.2% BSA prior to further use.
4. Add 20 µL of 50 µg/mL FITC- or phycoerythin (PE)-labeled antibodies and incu-
bate for 30 min in the dark at 4°C.
5. When epitopes on the membrane or in the CAM are analyzed (35,37), PI, which
binds to DNA in dead cells with an unintact plasma membrane, is added to recog-
nize and exclude dead cells. For DNA labeling, add 5 µL of PI solution/300 µL
cell suspension.
6. Wash cells with PBS before flow cytometer analysis.
3.5. Flow Cytometric Analysis

1. Stained cells are analyzed on a flow cytometer.
2. For each sample, 15,000–20,000 events are analyzed.
3. Cells are gated on forward and side scatter to exclude dead cells, debris, and
aggregates. Propidium iodide was additionally used to exclude dead cells when
the epitopes outside the cells and cell-associated ECM molecules were analyzed.
4. The mean fluorescence intensity (MFI) of the positive cell population, which is
owing to the binding of the conjugated antibodies to the specific antigen, is used to
quantify the presence of plasma membrane-associated epitopes, e.g., the receptors
for biological mediator proteins, the accumulation of the ECM molecules in the
CAM, and the levels of growth factors and cytokines, or distinct enzymes and their
natural inhibitors inside the cells (see Note 7).
5. MFI values are obtained by subtraction of the MFI of the negative control popu-
lation from the MFI of the positive stained population.
6. For comparison between experiments, Quantum Simply Cellular Microbead Kit
is used to calibrate the fluorescence scale of the flow cytometer (23).
7. The microbeads are stained and processed in parallel with the cell samples using
the same amount of FITC-labeled antibodies and incubation time.
8. The fluorescence scale of the cytometer is adapted before every experiment in
order to keep identical MFIs for the four peaks of the calibration beads.
9. The MFI of cell samples is then analyzed without changing any instrument set-
tings. The reproducibility and reliability of the whole procedure have been dem-
onstrated previously (35,37).
10. Classically, flow cytometry is used to define percentages of cells staining with
antigen-specific monoclonal antibodies. The interface channel for positivity is
arbitrarily set at the point at which less than 2% of the control fluorescence (cells
exposed to isotype-matched FITC-mouse IgG) is positive.
3.5.1. Reproducibility and Reliability Tests

To evaluate the reproducibility of the flow cytometry assays, the method
was tested on aliquots of cells obtained from the same culture. To study the
reliability of the whole procedure, cells from the same donor were cultured
separately in four cultures. Coefficients of variation were calculated to esti-
mate the reliability of the values obtained for the presence of the respective
cell-bound epitopes (35), by the formula:
Coefficient of variation = 1 SD × 100/X
where SD is standard deviation, and X is the mean of the values.
3.5.2. Influence of Isolation Procedures on Detection of Cell-Bound Molecules

1. In order to exclude the effects of the isolation methods on the presence of cell-
bound ECM molecules, isolated chondrocytes were analyzed by flow cytometry
before and after exposure to EDTA, Na-citrate, or agarase.
2. To test the effect of agarase, subconfluent monolayer cultured cells were de-
tached after 4 d of culture with 1 mmol/L EDTA in HBSS without Ca2+/Mg2+ and
harvested.
3. The expression of aggrecan and types I and II collagen was tested on the isolated
cells before and after exposure to 50 U/mL of agarase at 37°C for 1 h.
4. To test the effect of EDTA, chondrocytes were obtained from 1-wk-old agarose
cultures after digestion of the agarose gel with agarase.
5. The expression of aggrecan and types I and II collagen was then tested on the
isolated cells before and after exposure to EDTA in HBSS without Ca2+/Mg2+ (37).
3.6. Results and Discussion

3.6.1. Flow Cytometric Analysis
A typical flow cytometry dot plot is shown in Fig. 1. The chondrocytes in
this experiment were harvested after 1 wk of primary culture in agarose. A
two-parameter dot plot of the cells was performed following PI staining to
exclude the dead cells. According to their size (forward scatter) and their PI
staining, the cells separated on these two-dimensional plots into three groups:
PI-positive dead cells, PI-negative debris, and PI-negative living cells. The
population of single living cells was analyzed for their expression of the ECM
compounds. Flow cytometry histograms representing the proportions of
chondrocytes staining for aggrecan and types I and II collagen are shown in
Fig. 1 (35).
3.6.2. Reproducibility and Reliability of Flow Cytometry

To evaluate the reproducibility of the method, subconfluent monolayer cul-
tured cells were dissociated with EDTA and tested fourfold for the presence of
cell-bound ECM molecules. Coefficients of variation of the assay for aggrecan
and types I and II collagen were in between 2.5 and 13.2% (35). The same
procedure was repeated with a sample of third-passage monolayer-cultured
chondrocytes that were subcultured in agarose. Coefficients of variation of the
assay for the CAM ECM molecules were from 2.6 to 11.0% (35) (Table 2).
Table 2
Reproducibility (Interassay Variability)
of Flow Cytometric Analysis of Human Cartilage
Chondrocytes in Monolayer and in Agarose Culture a
X SD CV (%)
Monolayer
Aggrecan 66.2 2.5 3.8
Type II collagen 14.4 1.9 13.2
Type I collagen 25.9 0.6 2.5
Agarose
Aggrecan 14.7 0.9 6.1
a Percentages
of cells positive for a given epitope are represented.
X SD, mean 1 SD; CV, coefficient of variation = 1 SD × 100/X.
The reliability of the whole procedure was also tested on aliquots of four
chondrocyte monolayers originating from the same donor but cultured sepa-
rately (35). After 1 wk in culture, the subconfluent cells were dissociated with
EDTA and tested for the presence of cell-bound ECM molecules. Coefficients
of variation of the assay for aggrecan and types I and II collagen were 2.2,
18.8, and 9.0%, respectively. The same procedure was repeated with a sample
of second-passage monolayer-cultured chondrocytes that were subcultured in
four separate agarose cultures. After 1 wk, the agarose matrix was digested
with agarase to isolate the cells, which were tested for the presence of cell-
bound ECM epitopes. Coefficients of variation of the assay for aggrecan and
types I and II collagen ranged from 6.3 to 10.2% (35) (Table 3).
Fig. 1. (opposite page) Flow cytometric analysis of human chondrocytes harvested

after 1 wk of primary culture in gelled agarose. Top: Dot plot of human chondrocytes
after propidium iodide staining. R1, region 1 (PI-positive, dead cells); R2, region 2
(PI-negative, debris); R3, region 3 (PI-negative, living cells). Bottom: Histograms
showing the expression of (A) aggrecan, (B) type II collagen, (C) type I collagen.
Gray curves, negative controls; black curves, MAb FITC-labeled cells. In A, B, and C,
ordinates are cell counts, and abscissae are fluorescence intensity units.
Table 3
Reliability (Interculture Variability)
of Flow Cytometric Analysis of Human Cartilage
Chondrocytes in Monolayer and in Agarose Culture a
X SD CV (%)
Monolayer
Aggrecan 66.5 1.5 2.2
Agarose
Aggrecan 21.6 2.2 10.2
a Percentages of cells positive for a given epitope are represented.
X SD, mean 1 SD; CV, coefficient of variation = 1 SD × 100/X.
3.6.3. Influence of Isolation Methods on the Detection

of Cell-Bound ECM Molecules
To test whether or not treatment of the cells with EDTA altered the presence
of the cell-bound matrix antigens, chondrocytes cultured in agarose were released
from the gel with agarase. Proportions of cells recognized by anti-aggrecan and
antitypes I and II collagen MAbs did not change significantly after 30 min of
exposure to 1 mmol/L EDTA in HBSS without Ca2+/Mg2+ (35) (Table 4). To
test whether or not treatment of the cells with agarase altered the expression of
the cell-bound matrix antigens, monolayer-cultured chondrocytes were released
from the culture dishes with 1 mmol/L EDTA in HBSS without Ca2+/Mg2+ and
then treated with agarase at 37°C for 1 h. Proportions of cells recognised by anti-
aggrecan and antitypes I and II collagen MAbs did not change after exposure to
agarase (35) (Table 5).
3.6.4. Quantifying the Molecular Contents of the CAM
The availability of MAbs directed specifically against cell proteins has
enabled us to quantify numbers of intracellular and cell-associated molecules,
e.g., cytokines and growth factors, MMPs and their natural inhibitors inside
cells, and ECM proteins in the CAM (18,23,37,38). The fluorescence inten-
sity of the positive cell population, which is owing to the binding of FITC-
MAb with equimolar amounts of intracellular or cellbound molecules, can be
used to compare the quantities of these biologically active compounds, or to
quantify their absolute amounts in the cell or in the CAM (Table 6). When
Table 4
Chondrocyte Phenotype Before
and After Treatment With EDTA a
% Change of positive cells after EDTA
Aggrecan Collagen II Collagen I
M40 – 4.1 – 2.3 – 24.5
M25 – 3.6 + 0.3 + 26.2
F28 – 5.9 + 12.8 + 21.0
a Sex (M/F) and age (yr) of the donors are given. Primary
cultures were established in agarose, and the cells were har-
vested after 1 wk to test the effects of EDTA.
Table 5
Detection of Cell-Bound ECM Molecules
Before and After Digestion With Agarase a
% Change of positive cells after agarase
Aggrecan Collagen II Collagen I
M65 – 5.4 + 2.7 + 5.4
M40 + 7.1 + 4.2 – 1.3
M25 – 23.1 + 1.0 + 3.2
aSex (M/F) and age (yr) of the donors are given. Third-
passage monolayer cultures were established, and the cells
were harvested after 4 d to test the effects of agarase.
flow cytometry and enzyme-linked immunosorbent assay (ELISA) were used

to measure the aggrecan content in the CAM of a chondrocyte population,
both methods yielded equal results (Fig. 2) (37).
3.6.5. Chondrocyte Phenotype in the Monolayer
Condition and in Suspension Culture
The percentages of cells staining positively with MAbs against chondrocyte-
specific aggrecan and type II collagen in primary cultures in the monolayer
condition have been reported previously (35). The percentage of cells staining
for aggrecan and type II collagen decreased significantly during prolonged
monolayer culture, whereas the expression of type I collagen (which is a marker
of chondrocyte dedifferentiation) increased in the same culture condition in
primary cultures (Fig. 3A). This loss of the original phenotype became irre-
versible when secondary and tertiary cultures were investigated (35).
Table 6
Expression of IGFRI, IGF-1, IL-1RI, and IL-β
by Chondrocytes From Intact Cartilage From Healthy Joints a
Sex/age IGFRI IGF-1 IL-1RI IL-1β
M44 2.36 0.05 40.27 6.10 1.86 0.11 44.00 7.10

M47 4.25 0.15 42.20 0.55 2.55 0.18 60.64 1.98
M56 1.96 0.19 19.86 2.81 1.53 0.07 5.60 0.25
M59' 2.50 0.32 27.90 1.70 2.11 0.08 56.48 7.41
F40 5.41 0.15 44.50 3.85 1.57 0.08 36.27 3.20
M38 4.36 0.08 39.15 0.94 1.63 0.14 30.06 0.52
M18 2.64 0.10 27.03 0.26 1.75 0.05 56.05 0.05
M59'’ 1.82 0.05 21.96 4.30 2.98 0.48 72.90 4.56
M57 2.77 0.08 40.60 1.21 1.51 0.06 50.16 3.00
M18 0.95 0.09 21.03 3.00 0.91 0.21 6.39 0.56
F67 0.46 0.02 21.80 1.03 0.39 0.06 5.10 0.48
F73 0.47 0.03 32.10 0.47 0.24 0.03 12.20 0.84
Mean 1 SD b 2.50 1.55 31.53 9.47 1.59 0.80 36.32 24.1
a Sex (M/F) and age (yr) of the donors are given. Mean 1 SD MFI values for ligands and
their receptors are represented. As MAbs are used to stain protein epitopes, and equimolar
amounts of MAb and the targeted epitopes bind each other, MFI values stand for absolute amounts
of cell-associated proteins. The ligands outnumber their receptors by a factor of 10. IGF, insulin-
like growth factor; IGFRI, IGF receptor I; IL, interleukin; IL-1RI, IL-1 receptor I.
b Mean 1 SD MFI value recorded for the whole population (n = 12).
Re-expression of aggrecan and type II collagen by primary chondrocytes in

suspension culture in agarose was obvious when these cells were followed over
a 14-d period. However, synthesis of unusual type I collagen accompanies dif-
ferentiation of the articular cartilage cell in gelled agarose (Fig. 3B).
3.6.6. The Chondrocyte CAM Reflects the ECM of Articular Cartilage
Flow cytometry allowed homeostasis of the ECM to be studied. For instance,
exactly the same depression of aggrecan was obtained in the CAM and in the
interterritorial matrix (ITM) of the ECM, when chondrocytes had been exposed
to interleukin-1β (IL-1β). An equal recovery in CAM and ITM contents was
observed after the incubation media of these chondrocytes had been supple-
mented with increasing doses of IGF (Fig. 4) (18). Suppression of aggrecan in
the CAM and in the ITM was assayed by flow cytometry on the isolated
chondrocytes and by ELISA on the extract of the surrounding artificial matrix.
Fig. 2. Mean fluorescence intensity (MFI) owing to CAM aggrecan and ELISA
for aggrecan in the CAM was carried out after chondrocytes of two donors had been
exposed to IL-1. An identical procedure was used to prepare chondrocytes for flow
cytometry and for ELISA to quantify the CAM aggrecans. The anti-aggrecan MAb
used in flow cytometry was the same as the detection MAb used in the ELISA proce-
dure (Biosource). Analysis of the IL-1β-induced changes in CAM content of the
chondrocytes obtained in these two donors showed a close agreement between the
two methods, and a significant correlation was observed when the aggrecan MFI (flow
cytometry) and aggrecan absolute contents (ELISA) were compared under different
doses of IL-1β.
Fig. 3. Re-expression of cell membrane-bound extracellular matrix antigens by chon-

drocytes in secondary culture. (A) Monolayer-cultured chondrocytes, (B) chondrocytes
in gelled agarose. Ordinates, % of positive cells; abscissae, culture time in days.
Fig. 4. Dose-response effect of increasing doses of insulin-like growth factor-1

(IGF-1) on the accumulation of aggrecan in the cell-associated matrix (CAM) (A) and
interterritorial matrix (ITM) (B) of IL-1β-depressed chondrocytes. co, control cells
not exposed to IL-1. Mean values ± 1 SD for percentage changes of three experiments
in one chondrocyte population are shown.
3.6.7. Homeostasis of the ECM by Chondrocytes From Normal Cartilage

When addressing the homeostasis of the ECM by chondrocytes in normal
articular cartilage, our flow cytometry studies were focused on the accumula-
tion of aggrecan and type II collagen in the CAM of phenotypically stable
articular cartilage chondrocytes cultured in alginate.
Immediately after their isolation and after the initiation of the culture proce-
dure, the chondrocytes showed high levels of intracellular growth factors and
cytokines (18). Mechanical stress, sustained by the cells during the process of
their isolation, is thought to have initiated these metabolic changes. Similar
processes of activation of cells by mechanical forces, known as mechano-
transduction, have been described for different cells, e.g., skeletal muscle cells
(39), chondrocytes (40,41), and endothelial cells (42). However, elevated val-
Table 7
Correlations Among the Expression of IGFRI, IL-1RI, and IL-1RII
on the Cell Membrane, of IGF-1 and IL-1α and β Intracellularly
and of Aggrecan and Type II Collagen in the CAM a
IGFRI IGF-1 IL-1RI IL-1RII IL-1α IL-1β
IGF-1 S
IL-1RI — —
IL-1RII S S —
IL-1α — — S —
IL-1β — — S — S
Coll II S S — S — —
Aggr S S — S — —
a S, significant correlation; Coll II, type II collagen; Aggr, aggrecan; IGF-1, insulin-like growth
factor-1; IGFRI, IGF receptor I; IL, interleukin; IL-1R, IL-1 receptor.
ues of insulin-like growth factor-1 (IGF-1) and of IL-1α and -β inside the cells
decreased and stabilized within the first 2 wk in culture. The cells were thus
allowed to recover from the isolation procedure, to restore their repertoire of
plasma membrane receptor proteins, and to rebuild the cell-associated ECM
that had been lost during the isolation procedure. The new equilibriums were
reached after 1–2 wk in culture (18).
After 1 wk in culture, the cellular levels of both parts of the IGF receptor I
(IGFRI)/IGF-1 pathway significantly correlated with the amounts of aggrecan
and type II collagen that had accumulated in the CAM. In addition, there was a
significant correlation between IGFRI/IGF-1 and the presence of the decoy
receptor for IL-1:IL-1RII. Additionally, the same extent of correlation was found
between IL-1RII and the ECM molecules in the CAM. Levels of CAM ECM
molecules did not correlate with the agonists of the IL-1 pathway (Table 7) (18).
Cause/effect relation experiments were then performed to explore the
effects of IGF on synthesis and turnover of the ECM. IGF-1 was shown to dose-
dependently enhance the accumulation of aggrecan and of type II collagen in
the chondrocyte CAM (18). These results supported the observations of others
who have shown that growth factors, especially IGF, direct the production and
accumulation of ECM by chondrocytes in normal and diseased cartilage (43–
47). Furthermore, exogenous IGF-1 was shown to induce the expression of IL-
1RII on the chondrocyte plasma membrane. IL-1RII binds and neutralizes
IL-1β, but does not, or barely, bind or neutralize IL-1α in bioassays (48,49).
Fig. 5. Correlations observed between both insulin-like growth factor-1 (IGF-1)/

IGF receptor I (IGFRI) and interleukin-1 (IL-1)/IL-1 receptor (IL-1R) auto/paracrine
pathways and accumulated CAM compounds, and the results of a number of addi-
tional cause/effect experiments allowed us to conclude that IGF-1-induced IL-1RII
overrides IL-1 activity and controls the homeostasis of the extracellular matrix.
IL-1RII acts as a molecular trap for the IL-1 agonist without participating in its
signaling. Through the upregulation of IL-1RII, IGF-1 can protect cartilage
cell ECM against IL-1-induced destruction (Fig. 5). This was illustrated in our
in vitro experiments in which IGF-1 countered the action of IL-1β, e.g., the
deficient synthesis, degradation, and inadequate deposition of aggrecan in the
CAM and in the ECM of IL-1β-depressed chondrocytes (Fig. 4). This protec-
tive effect was shown to be modulated through the upregulation of plasma
membrane IL-1RII levels, since an IL-1RII-neutralizing IgG abolished the sup-
porting activity of IGF-1 (18). A decrease in both the basal and the cytokine-
stimulated degradation of proteoglycan by IGF-1 in cartilage explant cultures,
demonstrated earlier (47), is consistent with these findings.
3.6.8. Homeostasis of the ECM by Chondrocytes
From Osteoarthritic Cartilage
Although chondrocytes from degenerated tissues—when compared with
those from unaffected tissues from osteoarthritic (OA) joints—showed a sig-
nificantly increased expression of IGFR1 and IGF-1, equally significant
increases of intracellular IL-1α, IL-1β, and plasma membrane IL-1RI were
observed in these cartilage cells. On the other hand, chondrocytes derived from
degenerated tissue of OA joints expressed less membrane-bound IL-1RII than
cells isolated from unaffected cartilage of the same knee. It was anticipated that
IL-1 activity around chondrocytes from degenerated tissue was not neutralized
by IL-1RII and, therefore, mean chondrocyte MFI values for CAM aggrecan,
type II collagen, and fibronectin significantly decreased in these cells (50).
3.6.9. Use of Flow Cytometry to Assess the Results

of Pharmacological Interventions on Chondrocytes
Our studies have shown that chondrocytes treated with 0.05 µg/mL hydro-
cortisone showed a significantly increased accumulation of CAM aggrecan, type
II collagen, and fibronectin (38). This increased accumulation of CAM macro-
molecules obviously resulted from a decrease in the activity of the catabolic
pathways, since the intracellular levels of both IL-1 isoforms were depressed in
chondrocytes after the exposure to hydrocortisone. The intracellular growth fac-
tors and cytokines represent a reservoir of biologically active agents, of which
variable amounts are secreted following various stimuli. It can be anticipated
that the intracellular amounts of these bioactive molecules predict their extra-
cellular function. Further suppression of the IL-1 catabolic pathway was
observed as the expression of the signaling IL-1RI receptor decreased, and as
the plasma membrane IL-1RII decoy receptor levels raised concomitantly on
the cells exposed to hydrocortisone. The decreased activity of the IL-1–medi-
ated catabolic pathways in hydrocortisone-treated chondrocytes—as reported
by others (51)—explains the well-documented reduction of neutral protease
activities in steroid-treated OA cartilage samples (52–55).
3.7. Summary
Flow cytometry offers a technique to study intracellular and extracellular
cell-associated proteins. As far as the cells under investigation have survived
their isolation procedure without any degradation of their CAM- or plasma
membrane-associated molecular components, they can react with mono- or
polyclonal antibodies and be subjected to flow cytometry. The presence and
levels of intracellular proteins can be investigated in permeabilized cells.
Isolation of the chondrocytes from different culture conditions, e.g., mono-
layer culture or suspension culture in agarose or alginate gel, did not affect
their surrounding cell-bound matrix since treatment of the cells with EDTA,
agarase, or Na-citrate did not alter the expression of the cell-bound matrix
antigens. The flow cytometry technique was reproducible, and the whole pro-
cedure was shown to be reliable. Finally, assessment of the MFI of the chon-
drocytes offered a precise quantification of the synthesis and accumulation of
the ECM molecules studied.
Flow cytometry thus allows the composition of the CAM to be studied. The
biochemical composition of the CAM realistically reflects the instant pheno-
type of chondrocytes cultured in different culture conditions.
As the CAM represents a constant part of the ECM, the changes in the com-
position of the CAM closely reflect the ongoing metabolic processes in the
ECM. This methodology allows the influence of growth and differentiation
factors and cytokine cascades on the synthesis and degradation of a variety of
extracellular matrix components to be investigated. These investigations have

been done in cartilage cells obtained from normal and from OA tissues and
have allowed therapeutic targets in this disease to be defined.
The possibilities of pharmacological interventions, e.g., corticosteroids,
have been investigated.
4. Notes
1. In between the hyaluronidase/pronase and the collagenase digestion steps, it is
essential to build in an overnight recovery period for the chondrocytes in the
presence of serum. This improves viability of the cells during the entire 48-h
isolation procedure.
2. Adult human articular cartilage is particularly difficult to digest. The first
chondrocytes appear after 2–3 h of collagenase treatment. After 6 h, part of the
cartilage still remains undigested. If an attempt is made to digest these remaining
parts, additional exposure to collagenase will kill the cells that were liberated
after the initial 2–3 h.
3. Attachment and growth of the chondrocytes is better in Ham’s F12 with 1% PSF
and 10–20% FBS. For studies on synthesis and turnover of the ECM, DMEM
with 1% PSF and 10% FBS is preferred.
4. Proliferation of chondrocytes occurs at lower concentrations of the gel. At 1.5%
agarose, chondrocyte numbers during culture are fairly stable.
5. Culture media should be added with extreme care as the agarose gel is easily
disrupted when the nutrient media are added. As alginate beads are much more
stable in culture, replacement of the culture media can be done in no time.
6. Agarase, used to liberate chondrocytes from the agarose gel, is much more expen-
sive than the Na-citrate that is used to isolate chondrocytes from an alginate bead.
7. Mean or median fluorescence intensity (MFI) values characterize a chondrocyte
population much better than does the classically used % of positive cells on a
scattergram.
References
1.
1 Bell, R. S., Bouret, L. A., Bell, D. F., Gebhardt, M. C., Rosenberg, A., and Berrey,
H. B. et al. (1988) Evaluation of fluorescein diacetate for flow cytometric deter-
mination of cell viability in orthopaedic research. J. Orthop. Res. 6, 467–74.
2.
2 Vincent, F., Brun, H., Clain, E., Ronot, X., and Adolphe, M. (1989) Effects of
oxygen-free radicals on proliferation kinetics of cultured rabbit articular
chondrocytes. J. Cell Physiol. 141, 262–266.
3.
3 Jaffray, P., Ronot, X., Adolphe, M., Fontagne, J., and Lechat, P. (1984) Effects of
D-penicillamine on growth and cell cycle kinetics of cultured rabbit articular
chondrocytes. Ann. Rheum. Dis. 43, 333–338.
4.
4 Thuong-Guyot, M., Domarle, O., Pocidalo, J. J., and Hayem, G. (1994) Effects of
fluoroquinolones on cultured articular chondrocytes flow cytometric analysis of
free radical production. J. Pharmacol. Exp. Ther. 271, 1544–1549.
5.
5 Westacott, C. I., Atkins, R. M., Dieppe, P. A., and Elson, C. J. (1994) Tumor
necrosis factor-alpha receptor expression on chondrocytes isolated from human

articular cartilage. J. Rheumatol. 21, 1710–1715.
6.
6 Martel-Pelletier, J., McCollum, R., DiBattista, J., et al. (1992) The interleukin-1
receptor in normal and osteoarthritic human articular chondrocytes. Identification
as the type I receptor and analysis of binding kinetics and biologic function. Arthri-
tis Rheum. 35, 530–340.
7.
7 Lapadula, G., Iannione, F., Zuccaro, C., et al. (1998) Chondrocyte phenotyping in
human osteoarthritis. Clin. Rheumatol. 17, 99–104
8.
8 Bujia, J., Behrends, U., Rotter, N., Pitzke, P., Wilmes, E., and Hammer, C. (1996)
Expression of ICAM-1 on intact cartilage and isolated chondrocytes. In Vitro Cell.
Dev. Biol. Anim. 32, 116–122.
9.
9 Lance, E. M., Kimura, L. H., and Manibog, C. N. (1993) The expression of major
histocompatibility antigens on human articular chondrocytes. Clin. Orthop. 291,
266–282
10.
10 Lapadula, G., Iannione, F., Zuccaro, C., et al. (1995) Expression of membrane-
bound peptidases (CD10 and CD26) on human articular chondrocytes. Possible
role of neuropeptidases in the pathogenesis of osteoarthritis. Clin. Exp. Rheumatol.
13, 143–148.
11
11. Vivien, D., Boumedienne, K., Galera, P., Lebrun, E., and Pujol, J. P. (1992) Flow
cytometric detection of transforming growth factor-β expression in rabbit articu-
lar chondrocytes (RAC) in culture–association with S-phase traverse. Exp. Cell.
Res. 203, 56–61.
12.
12 Adolphe, M., Froger, B., Ronot, X., Corvol, M. T., and Forest, N. (1984) Cell
multiplication and type II collagen production by rabbit articular chondrocytes
cultivated in a defined medium. Exp. Cell. Res. 155, 527–536.
13. Mok, S. S., Masuda, K., Häuselmann, H. J., Aydelotte, M.B., and Thonar, E. J.-
M. A. (1994) Aggrecan synthesized by mature bovine chondrocytes suspended in
alginate. J. Biol. Chem. 269, 33,021–33,027.
14. Malfait, A. M. (1994) Development of an in vitro model to study cytokine-medi-
ated interactions between inflammatory cells and human chondrocytes. Ghent
University Thesis: pp. 45.
15.
15 Almqvist, K.F., Wang, L., Wang, J., et al. (2001) Culture of chondrocytes in algi-
nate surrounded by fibrin gel: characteristics of the cells over a period of 8 weeks.
Ann. Rheum. Dis. 60, 781–790.
16.
16 Platt, D., Wells, T., and Bayliss, M. T. (1997) Proteoglycan metabolism of equine
articular chondrocytes cultured in alginate beads. Res. Vet. Sci. 62, 39–47.
17. Bayliss, M. T. (1992) Metabolism of animal and human osteoarthritic cartilage, in
Articular Cartilage and Osteoarthritis. (Kuettner, K. E., Schleyerbach, R., Peyron,
J. G., Hascall, V. C., eds.), Raven, New York, NY, pp. 487–500.
18.
18 Wang, J., Elewaut, D., Veys, E. M., and Verbruggen, G. (2003) IGF-1-induced
IL-1RII overrules IL-1 activity and controls the homeostasis of the extracellular
matrix of cartilage. Arthritis Rheum. 48, 1281–1291.
19. Knudson, C. B. and Knudson, W. (1993) Hyaluronan-binding proteins in devel-
opment, tissue homeostasis, and disease. FASEB J. 7, 1231–1241.
20. Von der Mark, K., Mollenhauer, J., Pfaeffle, M., van Menxel, M., and Mueller, P.
K. (1986) Role of anchorin CII in the interaction of chondrocytes with extracellu-
lar collagen, in Articular Cartilage Biochemistry. (Kuettner, K.E., Schleyerbach,
R., and Hascall, V.C., eds.) Raven, New York, NY, pp.125–141.
21. Repraeger, A. and Bernfield, M. (1982) An integral membrane proteoglycan
can bind the extracellular matrix directly to the cytoskeleton. J. Cell Biol.
95(Abstract), A125.
22.
22 Hewitt, A. T., Varner, H. H., Silver, M. H., Dessau, W., Wilkes, C. M., and Mar-
tin, G.R. (1982) The isolation and partial characterization of chondronectin, an
attachment factor for chondrocytes. J. Biol. Chem. 257, 2330–2334.
23.
23 Wang, L., Almqvist, K. F., Veys, E. M., and Verbruggen, G. (2002) Control of
extracellular matrix homeostasis of normal cartilage by a TGFβ autocrine path-
way. Validation of flow cytometry as a tool to study chondrocyte metabolism in
vitro. Osteoarthritis Cartilage 10, 188–198.
24.
24 Wood, B. T., Thompson, S. H., and Goldstein, G. (1965) Fluorescent antibody
staining. III. Preparation of fluorescein-isothiocyanate-labeled antibodies.
J. Immunol. 95, 225–229.
25.
25 Kronick, M. N. and Grossman, P. D. (1983) Immunoassay techniques with fluo-
rescent phycobiliprotein conjugates. Clin. Chem. 29, 1582–1586.
26 Gysen, P. and Franchimont, P. (1984) Radioimmunoassay of proteoglycan.
26.
J. Immunoassay 5, 221–243.
27. Heinegård, D., and Oldberg, A. (1989) Structure and biology of cartilage and
bone matrix noncollagenous macromolecules. FASEB J. 3, 2042–2051.
28.
28 Kumagai, J., Sarkar, K., Uhthoff, H. K., Okawara, Y., and Coshima, A. (1994)
Immunohistochemical distribution of type I, II and III collagen in the rabbit
supraspinous tendon insertion. J. Anat. 185, 279–284.
29.
29 Green, W. T. (1971) Behavior of articular chondrocytes in cell culture. Clin.
Orthop. 75, 248–260.
30.
30 Kuettner, K. E., Pauli, B. U., Gall, G., McMemoli, V. A., and Schenk, R. K. (1982)
Synthesis of cartilage matrix by mammalian chondrocytes in vitro. Isolation, cul-
ture characteristics and morphology. J. Cell Biol. 93, 743–750.
31.
31 Benya, P. D. and Shaffer, J.D. (1982) Dedifferentiated chondrocytes reexpress
the differentiated collagen phenotype when cultured in agarose gels. Cell 30,
215–224.
32. Cornelissen, M., Verbruggen, G., Malfait, A. M., Veys, E. M., Broddelez, C., and
De Ridder, L. (1993) The study of representative populations of native aggrecan
aggregates synthesized by human chondrocytes in vitro. J. Tissue Culture Meth-
ods 15, 139–146.
33.
33 Verbruggen, G., Veys, E. M., Wieme, N., et al. (1990) The synthesis and
immobilisation of cartilage-specific proteoglycan by human chondrocytes in dif-
ferent concentrations of agarose. Clin. Exp. Rheumatol. 8, 371–378.
34. Guo, J., Jourdian, G.W., and MacCallum, D.K. (1989) Culture and growth
charecteristics of chondrocytes encapsulated in alginate beads. Connect. Tissue
Res. 9, 277–297.
35.
35 Wang, L., Verbruggen, G., Almqvist, K. F., Elewaut, D., Broddelez, C., and Veys,
E. M. (2001) Flow cytometric analysis of the human articular chondrocyte pheno-
type in vitro. Osteoarthritis Cartilage 9, 73–84.
36. Cornelissen, M., Dewulf, M., Verbruggen, G., et al. (1993) Size distribution of
native aggrecan aggregates of human articular chondrocytes in agarose. In Vitro
Cell. Dev. Biol. Anim. 29A, 356–358.
37.
37 Wang, L., Almqvist, K. F., Broddelez, C., Veys, E. M., and Verbruggen, G. (2001)
Evaluation of chondrocyte cell-associated matrix metabolism by flow cytometry.
38.
38 Wang, L., Wang, J., Almqvist, K. F., Veys, E. M., and Verbruggen, G. (2002) Influ-
ence of polysulphated polysaccharides and hydrocortisone on the extracellular
matrix metabolism of human articular chondrocytes in vitro. Clin. Exp. Rheumatol.
20, 669–676.
39.
39 Martineau, L. C. and Gardiner, P. F. (2001) Insight into skeletal muscle
mechanotransduction: MAPK activation is quantitatively related to tension.
J. Appl. Physiol. 91, 693–702.
40.
40 Hung, C. T., Henshaw, D. R., Wang, C. C., et al. (2000) Mitogen-activated pro-
tein kinase signaling in bovine articular chondrocytes in response to fluid flow
does not require calcium mobilization. J. Biomech. 33, 73–80.
41 Honda, K., Ohno, S., Tanimoto, K., et al. (2000) The effects of high magnitude
41.
cyclic tensile load on cartilage matrix metabolism in cultured chondrocytes. Eur.
J. Cell Biol. 79, 601–609.
42.
42 Fujiwara, K., Masuda, M., Osawa, M., Kano, Y., and Katoh, K. (2001) Is PECAM-
1 a mechanoresponsive molecule? Cell. Struct. Funct. 26, 11–17.
43.
43 Verschure, P. J., van Marle, J., Joosten, L. A., and van den Berg, W. B. (1995)
Chondrocyte IGF-1 receptor expression and responsiveness to IGF-1 stimulation
in mouse articular cartilage during various phases of experimentally induced
arthritis. Ann. Rheum. Dis. 54, 645–653.
44.
44 Guenther, H. L., Guenther, H. E., Froesch, E. R., and Fleisch, H. (1982) Effect of
insulin-like growth factor on collagen and glycosaminoglycan synthesis by rabbit
articular chondrocytes in culture. Experientia 38, 979–981.
45.
45 McQuillan, D. L., Handley, C. J., Campbell, M. A., Bolis, S., Milway, V. E., and
Herington, A. C. (1986) Stimulation of proteoglycan synthesis by serum and insu-
lin-like growth factor-1 in cultured bovine articular cartilage. Biochem. J. 240,
423–430.
46.
46 Tesch, G. H., Handley, C. J., Cornell, H. J., and Herington, A. C. (1992) Effects of
free and bound insulin-like growth factors on proteoglycan metabolism in articu-
lar cartilage explants. J. Orthop. Res. 10, 14–22.
47.
47 Tyler, J. A. (1989) Insulin-like growth factor 1 can decrease degradation and pro-
mote synthesis of proteoglycan in cartilage exposed to cytokines. Biochem. J.
260, 543–548.
48.
48 Kollewe, C., Neumann, D., and Martin, M. U. (2000) The first two N-terminal
immunoglobulin-like domains of soluble human IL-1 receptor type II are suffi-
cient to bind and neutralize IL-1β. FEBS Lett. 487, 189–193.
49.
49 Colotta, F., Saccani, S., Giri, J. G., et al. (1996) Regulated expression and release
of the IL-1 decoy receptor in human mononuclear phagocytes. J. Immunol. 156,
2534–2541.
50.
50 Wang, J., Verdonk, P., Elewaut, D., Veys, E. M., and Verbruggen, G. (2003) Homeo-
stasis of the extracellular matrix of normal and osteoarthritic human articular cartilage
chondrocytes in vitro. Osteoarthritis Cartilage 11, 801–809.
51.
51 Lee, S. W., Tsou, A. P., Chan, H., et al. (1988) Glucocorticoids selectively inhibit
the transcription of the interleukin 1β gene and decrease the stability of interleukin
1β mRNA. Proc. Natl. Acad. Sci. USA 85, 1204–1208.
52.
52 Pelletier, J. P., Mineau, F., Raynauld, J. P., Woessner, J. F., Gunja-Smith, Z., and
Martel-Pelletier, J. (1994) Intraarticular injections with methylprednisolone acetate
reduce osteoarthritic lesions in parallel with chondrocyte stromelysin synthesis in
experimental osteoarthritis. Arthritis Rheum. 37, 414–423.
53.
53 Pelletier, J. P., Martel-Pelletier, J., Cloutier, J. M., and Woessner, J. F. (1987)
Proteoglycan-degrading acid metalloproteinase activity in human osteoarthritic
cartilage, and the effects of intraarticular steroid injections. Arthritis Rheum. 30,
541–548.
54.
54 Pelletier, J. P. and Martel-Pelletier, J. (1985) Cartilage degradation by neutral
proteoglycanases in experimental osteoarthritis. Suppression by steroids. Arthri-
tis Rheum. 28, 1393–1401.
55. McGuire, M. B., Murhy, M., Reynolds, J. J., and Russel, R. G. G. (1981) Produc-
tion of collagenase and inhibitor (TIMP) by normal, rheumatoid and osteoarthritic
synovium in vitro: effects of hydrocortisone and indomethacine. Clin. Biol. 61,
703–710.
Proteoglycan Synthesis by Cultured Chondrocytes 209
15
A Simple and Reliable Assay of Proteoglycan
Synthesis by Cultured Chondrocytes
Frédéric De Ceuninck and Audrey Caliez
Summary
A simple and reliable method to measure proteoglycan synthesis by chondrocytes in cul-
ture is described. Confluent chondrocytes in 24-well plates are labeled for 24–72 h with
35SO 2– in the presence of stimulating agents. At the end of treatment, the secretion medium
4
containing radiolabeled neosynthesized secreted proteoglycans (SP) is harvested, and cell-
associated proteoglycans (CAP) are extracted with guanidine hydrochloride for 48 h. Aliquots
of medium and cell extracts are distributed on Whatman paper and left to dry. SP and CAP
are trapped by precipitation with the cationic detergent cetyl pyridinium chloride (CPC),
whereas nonincorporated 35SO 42– remains in solution. After drying, each spot corresponding
to one well in the plate is cut, and its radioactivity is measured. Counts are proportional to the
amount of neosynthesized proteoglycans. In a representative experiment using rabbit
chondrocytes, total proteoglycan synthesis (SP plus CAP) was increased 3.5-fold after the
addition of 1.34 nM insulin-like growth factor-1 (IGF-1) compared with nonstimulated cells.
Further addition of a fourfold molar ratio of IGF binding protein-3 completely abolished this
effect. This method can be used to measure proteoglycan synthesis by chondrocytes from
many species, including human osteoarthritic chondrocytes, as well as guinea pig, rabbit,
and rat chondrocytes.
Key Words: Chondrocyte; proteoglycan synthesis; glycosaminoglycan; anabolism; osteo-

arthritis; assay; cell culture; cetyl pyridinium chloride; precipitation, guanidine hydrochloride.
1. Introduction
Proteoglycans are, after collagens, the main solid components of articular car-
tilage. Cartilage proteoglycans are made of structurally different core proteins
bearing polysulfated glycosaminoglycan chains, which give cartilage its hydro-
philic nature and the osmotic pressure necessary to resist compressive loads. By
working on osteoarthritic chondrocytes, with impaired anabolic responses, or
more generally, on anabolic processes in cultured chondrocytes from different
209
210 De Ceuninck and Caliez
species, it is important to have an accurate measure of the overall anabolism. For

this, proteoglycans are a good choice since in addition to their biological impor-
tance in cartilage, their abundance totally eliminates potential problems of
method sensitivity.
The most abundant cartilage proteoglycan, aggrecan, contains a 250-kDa
core protein substituted with chondroitin sulfate and keratan sulfate chains and
forms a macromolecular complex of about 3000 kDa (1). The presence of the
chondroitin sulfate proteoglycan versican has also been found in articular car-
tilage, although its expression level is much lower than that of aggrecan (2).
Besides these two large aggregating proteoglycans, some smaller proteoglycans
belonging to the family of leucine-rich repeat (LRR) proteins are found in
articular cartilage. They include decorin and biglycan, which possess dermatan/
chondroitin sulfate chains (3), fibromodulin (4), and lumican (5), which pos-
sess keratan sulfate chains. These molecules have the ability to interact with
fibrillar collagens and are important for creating a functional network within
cartilage. A large variety of proteoglycans bearing heparan sulfate chains are
also synthesized by chondrocytes. These heparan sulfate proteoglycans include,
among others, the basement membrane heparan sulfate proteoglycan (HSPG)
perlecan (6,7), syndecans -1, -3, and -4 (8–10), the endocytic receptor for
hyaluronan, CD44 (11), and fibroglycan, glypican, and betaglycan (8). Some
methods designed to measure some of these proteoglycans specifically have
been developed. Generally, these assays use antibodies directed against
the core protein, which implies prior digestion of the glycosaminoglycan
chains to facilitate antibody accessibility. Alternatively, some dyes of varying
degree of specificity for glycosaminoglycan chains of proteoglycans have been
used to develop colorimetric assays. This chapter describes an assay that mea-
sures neosynthesized proteoglycans in their whole, irrespective of their struc-
tural diversity.
The method takes advantage of the fact that cartilage proteoglycans have a
high molar ratio of the sulphur atom per glycosaminoglycan chain, so that
labeling of chondrocytes with 35SO42– leads to quite exclusive incorporation of
the radioactive atom in neosynthesized proteoglycans. It is estimated that 35S
incorporated in cysteine and methionine of proteins is negligible, being less
than 5% of the total 35S taken by the cells. In the method, two pools consisting
of cell-associated and secreted neosynthesized 35S-labeled proteoglycans are
precipitated with the cationic detergent cetyl pyridinium chloride (CPC),
whereas free nonincorporated 35SO42– is eliminated by successive washes.
Simple counting in scintillation fluid gives an accurate measure of proteoglycan
synthesis by chondrocytes. Our routine method, originally developed in the
laboratory of Dr. M. Corvol in Paris, is described here in details, with tips to
improve sensitivity and accuracy.
2. Materials
2.1. Cell Culture and Treatment
1. Articular cartilage explants from guinea pigs, rabbits, or rats (for preparation, see
Chapter 16) or from human osteoarthritic cartilage obtained after surgery for total
knee joint replacement (femoral condyles and tibial plateaus) or total hip replace-
ment (femoral head).
2. Hanks’ balanced salt solution (HBSS; Gibco).
3. Ham’s F-12 culture medium with Glutamax (Gibco).
4. Fetal calf serum (FCS).
5. Penicillin (10,000 U/mL)/streptomycin (10,000 µg/mL) stock solution (PS;
Gibco).
6. Enzyme solutions: prepare a solution of 3 mg/mL of collagenase (type I,
Worthington) and 2 mg/mL of dispase (from Bacillus polymixa, Gibco) in either:
a. HBSS for digestion of cartilage from guinea pig, rabbit, or rat.
b. Ham’s F-12 supplemented with 10% FCS and 1% PS for digestion of human
cartilage.
7. Cell culture equipment including: 10-mm-diameter Petri dishes, pipets,
multipipets, and appropriate tips or combitips.
8. 24-Well culture plate, flat bottomed (Nunc).
9. Bovine serum albumin (BSA), IgG-free, protease-free (Jackson Immuno
Research). Prepare a stock solution of 10% BSA (w/v) in deionized water and
filter through disposable units fitted with 0.22-µm filters. Freeze in aliquots of
1 mL until use.
10. Insulin-like growth factor-1 (IGF-1); insulin-like growth factor binding protein-
3 (IGFBP-3; Sigma).
11. Sulphur-35, sulfate, 100 µCi (3.7 MBq) in aqueous solution (35SO42–; Amersham).
2.2. Proteoglycan Precipitation and Extraction

1. Dulbecco’s phosphate-buffered saline without calcium and magnesium (DPBS;
Gibco).
2. Guanidine hydrochloride (Fluka).
3. Cetyl pyridinium chloride (CPC; Sigma).
4. Proteoglycan extraction solution: 3 M guanidine hydrochloride, 50 mM Tris-HCl,
pH 7.4. Store at 4°C.
5. Whatman chromatography paper, 3 MM. For one culture plate in the experiment,
cut two rectangles of exactly 12 × 18 cm. One will be used for secreted
proteoglycans and the other for cell-associated proteoglycans. On each rectangle,
draw 24 squares of 9 cm2 with a lead pencil, reproducing the pattern of a 24-well
plate. Number each square in one of its angles (see Note 1).
6. Trays of 15 × 21 cm (ideally) or more.
7. Precipitating solution: 1% (w/v) CPC containing 0.3 M NaCl. Do not adjust pH
(see Notes 2 and 3). Prepare fresh just before use.
8. Rocking platform.
9. Mixer, of Vortexer type.

10. Safe-lock microtubes of 1.5 mL (Eppendorf or equivalent).
11. Scintillation fluid.
12. Safety equipment for radioactive material.
13. Additional tools: gloves, scissors, lead pencil.
3. Methods
3.1. Chondrocyte Culture and Treatment
3.1.1. Chondrocyte Culture
Chondrocytes from any species can be isolated and cultured as described in
Chapter 1 (see Note 4). This alternative method is routinely used in our laboratory:
1. Finely mince the cartilage isolated from animals (guinea pig, rat, or rabbit) or
humans down to fragments of 1-mm size.
2. Transfer the fragments to a Petri dish containing 20 mL of the appropriate enzyme
solution, as described in Subheading 2.1., item 6.
3. Incubate at 37°C under a 5% CO2 atmosphere. For cartilage from guinea pig, rat,
or rabbit, digestion time is 5–6 h. For human cartilage, digestion time is 16 h (see
Note 5).
4. Collect cells by centrifugation (3 min at 900g). Resuspend in Ham’s F-12 medium
containing Glutamax plus 10% FCS and 1% PS. Count cells and dispense in
24-well plates at 105 cells/mL/well (see Note 6).
5. Change medium every 2–3 d until chondrocytes are confluent (see Note 7).
6. Replace culture medium with 1.2 mL/well of serum-free Ham’s F-12 medium
containing BSA at a final concentration of 0.1% (w/v; stock solution diluted
1:100) for 24 h.
3.1.2. Treatment (see Note 4)

1. Replace medium with 400 µL (treated wells) or 450 µL (control wells) of fresh
serum-free medium containing 0.1% BSA. In treated wells, add 50 µL of growth
factor or cytokine to be studied at the desired concentration. Finally, add in every
well 50 µL of a 35SO42– solution at 15 µCi/mL (final concentration 1.5 µCi/mL).
Each condition should have at least four replicates to ensure reproducibility and
to allow statistical analysis of the results.
2. Treatments may last for 24, 48, or 72 h, as required (see Note 8).
3.2. Measure of Proteoglycan Synthesis

The overall method is depicted in Fig. 1 and described here in detail.
1. At the end of treatment, harvest the secretion media in 1.5-mL-safe-lock
microtubes and centrifuge for 5 min at 900g at 4°C to pellet floating cells. For
further processing of media, go to step 5.
2. Wash the cells three times with 1 mL of DPBS to eliminate residual 35SO42– not
specifically associated to chondrocytes.
Fig 1. Schematic representation of the method for measuring proteoglycan synthe-

sis by chondrocytes. CPC, cetyl pyridinium chloride.
3. Dispense 0.5 mL of proteoglycan extraction solution in each well and rock the
plates gently for 48 h at 4°C to ensure complete extraction of cell-associated
proteoglycans (CAP).
4. Harvest extraction mixtures in 1.5-mL safe-lock microtubes and centrifuge for
10 min at 10,000g. The supernatant contains CAP.
5. The supernatant (collected as in step 1) containing secreted proteoglycans (SP)

and the fraction containing CAP (as collected in step 4) are then processed in a
similar way.
6. Take 50 µL of the solution of either SP or CAP from the microtube and lay exactly
at the center of a 9-cm2 square in Whatman 3MM predrawn sheets (see Note 9).
Perform this operation on separate sheets, one for SP and one for CAP.
7. Let dry for at least 2 h at room temperature.
8. Plunge the dry sheets into 15 × 21 cm trays (one sheet per tray) containing 100 mL
of precipitating solution (see Note 3) and shake gently for 30 min at room tempera-
ture. Repeat three times by renewing with fresh precipitating solution. During this
step, precipitated 35SO 42– radiolabeled proteoglycans are trapped within the
Whatman sheets, whereas free unincorporated 35SO42– stays in solution.
9. Let dry for 24 h at room temperature, or 5 h at 37°C (ensure that the sheets are
completely dry).
10. Cut each 9-cm2 square, roll it (see Note 2) and put it into a scintillation vial.
11. Add scintillation fluid and count (see Note 10).
3.3. Expression of Results

3.3.1. Total Proteoglycan Synthesis
1. Assuming that in Fig. 1 (upper left), A are four controls (untreated wells), and B
are four treated wells, the effect of treatment in terms of percent total proteoglycan
synthesis compared with controls can be expressed as [(mean dpm of SP + mean
dpm of CAP) in B]/[(mean dpm of SP + mean dpm of CAP) in A] × 100.
2. A typical experiment is depicted in Fig. 2, where IGF-I at 1.34 nM stimulated the
basal synthesis of proteoglycans by rabbit chondrocytes approx 3.5 times after a
24-h treatment. Cotreatment of chondrocytes with IGF-I and a fourfold molar
excess of IGFBP-3, one of the binding proteins that captures IGF-I and neutral-
izes its biological effects, brought back proteoglycan synthesis to the control
value. Note that the high level of precision of the method leads to very low intra-
assay variation, with standard errors typically being around 2% of the means.
3. The basal incorporation of 35SO42– by chondrocytes depends on the species stud-
ied. Table 1 shows that, with 1.66 × 106 dpm added at T0, about seven times
higher amounts of 35SO42– were incorporated into proteoglycans neosynthesized
by rabbit chondrocytes than by human osteoarthritic chondrocytes. According to
this observation, the counting time should be adjusted to obtain maximum accu-
racy (see Note 10).
3.3.2. Cell-Associated Proteoglycans

1. Depending on the conditions, some factors may affect the distribution of
neosynthesized proteoglycans between the cells and the culture medium. In this
case, the percentage of cell-associated proteoglycans/total proteoglycans for
treated cells compared with control cells is calculated as: [mean dpm of CAP/
(mean dpm of SP + mean dpm of CAP) in B]/[mean dpm of CAP/(mean dpm of
SP + mean dpm of CAP) in A] × 100.
Fig 2. Regulation of total proteoglycan synthesis (SP + CAP) by actors of the IGF
system in cultured rabbit chondrocytes. Shown is one representative experiment out of
six. Bars are means ± SEM of four replicates for each condition. Statistical differ-
ences: insulin-like growth factor-1 (IGF-1) vs control, p < 0.001; IGF binding protein-
3 (IGFBP-3) vs IGF-1, p < 0.001.
Table 1
Relative Proteoglycan Synthesis by Chondrocytes
From Different Species After 24 h of Incubation With 35SO42–
Species No. dpm
of chondrocytes of experiments (mean SEM)
Human 6 4120 883
Rabbit 6 28,300 7360
Guinea pig 3 11,100 2980
Rat 1 13,000
2. The proportion of CAP may vary when working with chondrocytes isolated
from different species. For example, the percentage of CAP in nonstimulated
rabbit chondrocytes after 24 h of incubation with 35SO42– was 32.4 3.2%
(mean SEM of six independent experiments), and was only 10.9 0.5%
(mean SEM of five independent experiments) in human osteoarthritic
chondrocytes in similar conditions.
4. Notes
1. Lines and numbers should be drawn exclusively with a lead pencil. Any other
tools such as ball-point pens or felt pens should be avoided, since the writing will
disappear when the sheets are plunged into the precipitation baths.
2. Caution: care should be taken when handling CPC since the powder is very fine
and light, and its inhalation may cause lung irritation. This is also valid when roll-
ing the dried 3 × 3-cm Whatman squares at the end of the experiment (see Sub-
heading 3.2.10.), since fine particles may flutter around. Always wear mask,
gloves, and glasses when using CPC!
3. The solution does not mix (or precipitates) at a temperature below 24°C, so it is
necessary to heat and maintain it at a higher temperature (up to 37°C). With respect
to this point, keep in mind that if the solution is not warm enough it could precipi-
tate during the baths, and this could lead to uneven precipitation of proteoglycans
within the Whatman sheets.
4. During the culture of cells and more especially during the treatment, it is important
to use a medium containing low amounts of sulfate salts (MgSO4, ZnSO4, CuSO4,
or FeSO4). The assay rests on the incorporation of exogenous 35SO42– added at
relatively low amounts, so too high amounts of endogenous nonradioactive sulfate
would decrease the incorporation of 35SO42– and thus the precision of the assay.
For this reason, Ham’s F-12 medium is preferred over DMEM medium, since the
sulfate concentration of the former is about 100-fold lower than for the latter.
5. Starting from an amount of human cartilage explants similar to that obtained
from one joint of small animals (and keeping the same volume/cartilage weight
ratio of the digestion mixture), it will require a longer time to achieve complete
digestion of human material. To ensure chondrocyte viability, the digestion is
performed in nutritive culture medium with FCS, although, on the other hand, the
presence of endogenous collagenase inhibitors in FCS may somewhat lower the
digestion. To optimize digestion, it is strongly advisable to divide the pool of
explants in several Petri dishes depending on the number of human explants, as
enzymes may run out if the ratio of tissue/enzyme is too high.
6. For a similar weight of explants before digestion, the number of cells obtained
will be about twofold lower for human cartilage explants compared with other
species and is highly variable depending on the age of donor and stage of osteoar-
thritis.
7. Human osteoarthritic chondrocytes usually have a higher doubling time compared
with chondrocytes from other species. Typically, the former will reach confluence
in about 2 wk instead of about 1 wk for the latter.
8. In this method, it is not mandatory to normalize the amount of proteoglycans to the
number of cells or DNA content, provided that the cells are confluent and that the
treatment lasts for 24 or 48 h. Indeed, confluent chondrocytes from most species
remain in monolayers for this time, and a treatment with growth factors such as
IGF-I or TGF-β do not affect cell division. In this case, “normalization” would
increase the variability of the results. It should be verified by every bench scientist
whether or not the cell number is affected, depending on the factors and the times
studied.
9. Touch the Whatman paper lightly with the end of the tip and dispense liquid on the
square so that it enters the sheet slowly by capillarity. Try to aim accurately at the
center of the square, to avoid contaminating the adjacent squares, since 9-cm2
squares hold just 50 mL of medium (leaving just a few millimeters dry at the sides
and the angles).
10. Actually, we fix parameters on the beta counter so that counting of a vial stops
when 2 sigma (2s) reaches 2%. This value represents the percent of uncertainty in
a gross count value (with 95% confidence limits). It is calculated as:
% 2s = 200/ accumulated counts

For uncertainty 2s = 2%, total accumulated counts should be 10,000. Thus, the
precision of the counting is independent of the number of dpm.
References
1.
1 Kiani, C., Chen, L., Wu, Y. J., and Yang, B. B. (2002) Structure and function of
aggrecan. Cell Res. 12, 19–32.
2.
2 Grover, J. and Roughley, P. J. (1993) Versican gene expression in human articu-
lar cartilage and comparison of mRNA splicing variation with aggrecan. Biochem.
J. 291, 361–367.
3.
3 Rosenberg, L. C., Choi, H. U., Tang, L. H., et al. (1985) Isolation of dermatan
sulfate proteoglycans from mature bovine articular cartilages. J. Biol. Chem. 260,
6304–6313.
4. Heinegard, D., Larsson, T., Sommarin, Y., Franzen, A., Paulsson, M., and Hedbom,
E. (1986) Two novel matrix proteins isolated from articular cartilage show wide
distributions among connective tissues. J. Biol. Chem. 261, 13,866–13,872.
5. Grover, J., Chen, X. N., Korenberg, J. R., and Roughley, P. J. (1995) The human
lumican gene. Organization, chromosomal location, and expression in articular
cartilage. J. Biol. Chem. 270, 21,942–21,949.
6.
6 SundarRaj, N., Fite, D., Ledbetter, S., Chakravarti, S., and Hassell, J. R. (1995)
Perlecan is a component of cartilage matrix and promotes chondrocyte attach-
ment. J. Cell Sci. 108, 2663–2672.
7.
7 Costell, M., Gustaffson, E., Aszodi, A., et al. (1999) Perlecan maintains the integ-
rity of cartilage and some basement membranes. J. Cell Biol. 147, 1109–1122.
8.
8 Grover, J. and Roughley, P., J. (1995) Expression of cell-surface proteoglycan
mRNA by human articular chondrocytes. Biochem. J. 309, 963–968.
9.
9 Barre, P. E., Redini, F., Boumediene, K., Vielpeau, C., and Pujol, J. P. (2000)
Semiquantitative reverse transcription-polymerase chain reaction analysis of
syndecan-1 and -4 messages in cartilage and cultured chondrocytes from osteoar-
thritic joints. Osteoarthritis Cartilage 8, 34–43.
10.
10 Pfander, D., Swoboda, B., and Kirsch, T. (2001) Expression of early and late
differentiation markers (proliferating cell nuclear antigen, syndecan-3, annexin
VI, and alkaline phosphatase) by human osteoarthritic chondrocytes. Am. J.
Pathol. 159, 1777–1783.
11. Hua, Q., Knudson, C. D., and Knudson, W. (1993) Internalization of hyaluronan by
chondrocytes occurs via receptor-mediated endocytosis. J. Cell Sci. 106, 365–375.
Assays of Cartilage Degradation 219
16
Assays of Proteoglycan and Collagen Degradation
in Cultures of Rabbit Cartilage Explants
Christophe Lesur and Massimo Sabatini
Summary
Cultures of cartilage explants have long been used to study the effects of modulators of
extracellular matrix degradation. We present a simple and rapid assay system, based on culture
of rabbit cartilage explants, which permits study of the effects of protease inhibitors on
proteoglycan degradation (caused by either aggrecanases or matrix metalloproteinases
[MMPs]), and on collagen degradation. The assay is based on the ability of interleukin-1 to
stimulate both aggrecanase activity and synthesis of inactive MMPs, which are then activated
by p-aminophenylmercuric acetate for the study of MMP-mediated proteoglycan degradation
or by plasmin for the study of collagen degradation. Proteoglycan degradation is quantified as
percent release of radioactivity from cartilage explants previously labeled with 35SO42–. Col-
lagen degradation is calculated as percent release of collagen, measured by colorimetric assay
of hydroxyproline.
Key Words: Cartilage; degradation; proteoglycan; aggrecan; aggrecanases; ADAMTS; col-

lagen; matrix metalloproteinases; hydroxyproline.
1. Introduction
Cartilage is made of chondrocytes embedded in an abundant extracellular
matrix (ECM), which accounts for about 95% of the volume of the tissue. The
two major components of the ECM are collagen, mainly type II, which forms a
dense network responsible for tensile strength, and aggrecan, a highly hydro-
philic proteoglycan responsible for resistance of cartilage to compression.
Arthritic diseases such as osteoarthritis and rheumatoid arthritis are associated
with excessive degradation of the ECM, which is the result of combined action
of several enzymes, mainly aggrecanases and matrix metalloproteinases
(MMPs) (1). Cultures of cartilage explants have long been used to study the
effects of modulators of ECM degradation. Better knowledge of the enzyme
219
220 Lesur and Sabatini
families involved in cartilage loss made possible simple adjustments of already

existing culture protocols, in order to separately analyze the degradative
activities of aggrecanases and MMPs.
We present here a simple assay system, based on culture of rabbit cartilage
explants, that allows study of the effects of protease inhibitors on the three
major aspects of cartilage loss, namely, aggrecanase- and MMP-mediated
proteoglycan degradation, and collagen degradation.
1.1. Degradation of Proteoglycan
Aggrecan, the most abundant cartilage proteoglycan, is a large macromol-
ecule (about 2500 kDa) made up of a core protein (about 250 kDa) to which
keratan and chondroitin sulfate glycosaminoglycans (GAGs) are linked (2).
Aggrecan core protein can be degraded by both aggrecanases and MMPs. Two
aggrecanases were recently identified as members of the ADAMTS family
(a disintegrin and metalloproteinase with thrombospondin domain), namely,
aggrecanase-1/ADAMTS-4 and aggrecanase-2/ADAMTS-5 (3,4). The effects
of aggrecanases and MMPs on aggrecan can be classically distinguished on the
basis of different cleavage sites in the interglobular domain. Whereas
aggrecanases cleave at NITEGE373—374ARGSVIL, MMPs cleave at VDIPEN341
—342FFGV. The different terminal neoepitopes can be recognized using
specific antibodies, allowing discrimination of aggrecan degradative activity
(Fig. 1) (5).
We previously determined in rat cartilage explants, using Western blot
analysis of NITEGE and VDIPEN in tissue extracts, the conditions under which
proteoglycans are degraded in vitro by either aggrecanases or MMPs (6).
Proteoglycan degradation was quantified as percent release of radioactivity
from cartilage explants previously labeled with 35SO42–. We tried the same
protocol with rabbit cartilage explants and found a stronger response to
interleukin-1 (IL-1), in terms of both aggrecanase and MMP degradative
activities, coupled with lower dispersion of the data, and easier collection of
the tissue from larger joints.
The assay is based on the ability of IL-1 to stimulate both aggrecanase activ-
ity and pro-MMP production. Since no endogenous activation of pro-MMPs
takes place in rabbit cartilage explants, IL-1-induced proteoglycan degradation
depends on aggrecanases only, as shown by increased levels of NITEGE, but
not VDIPEN (Fig. 2A). If, instead, IL-1-induced pro-MMPs are activated by
p-aminophenylmercuric acetate (APMA), then the breakdown of aggrecan is
mediated by MMPs, as shown by a strong increase of VDIPEN (Fig. 2B).
To confirm the specificity of aggrecan degradation caused by aggrecanases or
MMPs, we used AG-3340 (prinomastat), which inhibits several MMPs with Ki
values around 10–10 M (7), and also inhibits aggrecanases at concentrations
Fig. 1. General structure of the aggrecan molecule and cleavage sites in the G1–G2
interglobular domain of the core protein.
around 10–6 M (8). In IL-1-stimulated cartilage explants, 10–5 M AG-3340

blocked the increase of NITEGE levels in tissue extracts (Fig. 2A), as well as
proteoglycan degradation, measured as radioactivity release, with a mean inhibi-
tory concentration (IC50) of 7 × 10–7 M (Fig. 2C). In the case of proteoglycan
degradation induced by IL-1 followed by APMA, 10–7 M AG-3340 was suffi-
cient to block both the increase of VDIPEN (Fig. 2B) and proteoglycan release,
with an IC50 of 2 × 10–9 M (Fig. 2D). Different sensitivities to AG-3340 confirm
that proteoglycan degradation owing to IL-1, or IL-1 followed by APMA, is
caused by different types of enzyme.
1.2. Degradation of Collagen
Type II collagen represents 90–95% of the total collagen content of the carti-
lage, and is found in fibrils made of type II collagen molecules stabilized by
collagen IX and XI. Each type II collagen molecule is composed of a triple helix
of three identical α-chains covalently linked at their ends (9). Native fibrillar
collagen is known to be highly resistant to proteases, and its initial cleavage must
be collagenase-mediated. Three collagenases have been identified, MMP-1,
MMP-8, and MMP-13, which cleave type II collagen at a single point, three-
fourths of the distance from the N terminus (Fig. 3).
222
Fig. 3. Schematic representation of type II collagen and matrix metalloproteinase

(MMP) cleavage site.
Once denatured by collagenases, type II collagen is further degraded by

gelatinases (MMP-2 and MMP-9) and stromelysin (MMP-3) (10). Collagen
degradation in cultures of rabbit cartilage explants can be stimulated by the
addition of IL-1 and plasminogen (11). The cytokine increases both plasmi-
nogen activation and the production of pro-MMPs, which are in turn activated
by plasmin. Plasminogen probably plays an important role in pathological
activation of pro-MMPs, since it is present in the synovial fluid of patients
with osteoarthritis and rheumatoid arthritis (12). Here we describe a modifi-
cation of the assay by Saito et al. (11), which allows for a more rapid MMP
activation and ECM degradation, mainly by the replacement of plasminogen
with plasmin.
Collagen degradation is calculated as percent release of collagen, mea-
sured by colorimetric assay of hydroxyproline (OH-pro) in medium and tis-
sue hydrolysates, using a modification of Grant’s method (13). Using this
protocol, the MMP inhibitor blocked collagen degradation at 10–7 M with an
IC50 of 4 × 10–8 M (Fig. 4).
Fig. 2. (opposite page) Effects of interleukin-1 (IL-1) alone and IL-1 followed by
p-aminophenylmercuric acetate (APMA) on NITEGE and VDIPEN levels (A,B) and
proteoglycan degradation (C,D) in the absence and presence of AG-3340 in rabbit
cartilage explant cultures. Plans of the experiments are shown at the top of the figure.
(A) Effect of IL-1 on NITEGE and VDIPEN levels, analyzed by Western blot, in the
absence and presence of AG-3340. (B) Effect of IL-1 followed by APMA, on VDIPEN
levels, analyzed by Western blot, in the absence and presence of AG-3340. (C) Effect
of IL-1 on proteoglycan degradation in the absence and presence of AG-3340.
Proteoglycan degradation was measured as % release of radiolabeled material between
d 0 and 2. Asterisks indicate a significant difference between IL-1 alone and IL-1 plus
AG-3340: ***, p < 0.001, **, p < 0.01, *, p < 0.5; data are averages ± Sem; n = 8. (D)
Effect of IL-1 followed by APMA on proteoglycan degradation in the absence and
presence of AG-3340. Proteoglycan degradation was measured as % release of radio-
labeled material between d 1 and 2. Asterisks indicate a significant difference between
APMA alone and APMA plus AG-3340: ***, p < 0.001; data are averages ± Sem; n = 8.
Fig. 4. Effect of interleukin-1 (IL-1) plus plasmin on collagen degradation in the

absence and presence of AG-3340 after 2 d of treatment. Collagen degradation was
measured as % release of OH-pro. Asterisks indicate a significant difference between
IL-1/plasmin alone and IL-1/plasmin plus AG-3340: ***, p < 0.001, **, p < 0.01; data
are averages ± Sem; n = 8.
2. Materials
2.1. Cartilage Explant Preparation and Culture
1. Cartilage source: male New Zealand albino rabbits weighing 500–600 g.
2. Sterile surgical instruments: large (16 cm) and small (12 cm) scissors, large and
small tissue forceps, fine, curved pliers.
3. Sterile disposable scalpels with straight (no. 11) and curved (no. 23) blades
(Feather).
4. Dulbecco’s phosphate-buffered saline without calcium and magnesium (PBS;
Gibco).
5. Dulbecco’s minimal essential medium/Ham’s F12 medium, 50/50 mixture with
Glutamax™ (DMEM/F12; Gibco).
6. Penicillin (10,000 U/mL) plus streptomycin (10 mg/mL), 100X concentrated

stock solution (PS; Gibco).
7. Fetal bovine serum (FBS), previously decomplemented by heating for 30 min at
56°C (Gibco).
8. Bovine serum albumin (BSA), 35%, fraction V solution (Sigma).
9. Recombinant mouse IL-1β (Sigma): make up a stock solution at 5 µg/mL in PBS
containing 0.1% BSA, aliquot and store at –20°C.
10. APMA (Sigma): make up a 100X concentrated solution at 0.05 M in 0.05 M
sodium hydroxide. Prepare fresh on day of use (see Note 1).
11. Dimethyl sulfoxide (DMSO): used as solvent for tested compounds.
12. Cell culture dish, 100 × 20 mm (Becton Dickinson).
13. Sterile 5-mL polypropylene round-bottomed test tubes (Becton Dickinson).
14. 96-Well culture plate, flat bottomed (Becton Dickinson).
15. Sterile laminar flow hood for tissue culture.
16. Tissue culture incubator, with humidified atmosphere of 5% CO2/air at 37°C.
2.2. Cartilage Degradation (Aggrecan; 35SO42–)

1. DMEM/F12 medium supplemented with 10% FBS and 1% PS: to 500 mL of
DMEM/F12 add 55 mL of FBS and 5 mL of PS (10% FBS DMEM/F12).
2. DMEM/F12 medium supplemented with 0.1% BSA and 1% PS: to 500 mL of
DMEM/F12 add 1.43 mL of BSA 35% and 5 mL of PS (0.1% BSA DMEM/
F12).
3. Sulphur 35, sulfate, in aqueous solution (35SO42–); (Amersham): use at a final
activity of 3.7 MBq (100 µCi) for explant labeling. Use appropriate protective
equipment (gloves, goggles, and plexiglass screen) when handling the source and
labeled culture material.
4. DL-Dithiothreitol (DTT; Sigma).
5. EDTA (Sigma).
6. 20 mM Phosphate buffer: 2.76 g/L NaH2PO4·H2O, 0.38 g/L EDTA (1 mM), pH 6.8.
7. Papain (Sigma): make up at 0.6 mg/mL (w/v) in 20 mM phosphate buffer plus
0.25 mg/mL (w/v) DTT. Prepare fresh on day of use.
8. Miniature 6-mL vial (Pony Vial™; Packard).
9. Cocktail for liquid scintillation (LSC; Ultima Gold; Packard).
10. Liquid scintillation analyzer (Tri Carb 2900 TR; Packard).
11. Oven, set at 56°C, with timer.
2.3. Cartilage Degradation (Collagen; OH-pro)

In the following steps, handle the corrosive solutions (hydrochloric acid,
perchloric acid and sodium hydroxide) under a hood, with appropriate protec-
tive equipment (gloves, goggles).
1. DMEM/F12 medium supplemented with 0.1% BSA and 1% PS: to 500 mL of
DMEM/F12 add 1.43 mL of BSA 35% and 5 mL of PS 100X concentrated (0.1%
BSA DMEM/F12).
2. Plasmin, ε-aminocaproic acid (EACA)- and lysine-free from human plasma

(Calbiochem): make up a stock solution at 10 U/mL in sterile deionized water,
aliquot, and store at –20°C.
3. Hydrochloric acid fuming 37% (HCl 12 M; Merck).
4. Perchloric acid 70–72% (Merck).
5. 6 M Sodium hydroxide solution (NaOH 6 M): 60 g/250 mL final vol, in deionized
water. Solution heats strongly during solubilization.
6. OH-pro (Merck): for standard curve preparation, prepare a stock solution at
5 mM in deionized water, aliquot and store at –20°C.
7. OH-pro buffer: 30 g citric acid monohydrate, 15 g sodium hydroxide, 90 g sodium
acetate anhydrous; adjust to pH 6.0 and complete to 1000 mL with deionized
water.
8. Chloramine T trihydrate (CT; Merck): 96 mg of CT, 6.2 mL OH-pro buffer,
13.8 mL deionized water (sufficient for one 96-well plate). Prepare fresh on day
of use.
9. 4-(Dimethylamino)benzaldehyde (pDAB; Merck): to 1.47 g of pDAB, add
7.4 mL deionized water and 2.6 mL perchloric acid (sufficient for one 96-well
plate). Prepare fresh on day of use.
10. Borosilicate vials, 2 mL with cap (Wheaton).
11. 96-Well deep plate, 1.2 mL/well capacity (Costar).
12. 96-Well microtiter plate, flat bottomed (Costar).
13. Photometric microplate reader with 540-nm filter and calculation software
(Labsystems iEMS reader and Genesis™ software).
14. Oven, set at 110°C, with timer.
15. Water thermostated bath, set at 70°C.
3. Methods
3.1. Cartilage Explant Preparation
3.1.1. Animal Sacrifice
Anesthetize rabbits with isoflurane (Forene®, Abbott) using standard veteri-
nary equipment (Minerve®, France) (Fig. 5A), then sacrifice by rupture of the
cervical vertebrae, and exsanguinate by carotid section. The use of animals
must conform to ethical guidelines. This procedure can only be used after
acceptance by the local Ethics Committee on animal experimentation.
3.1.2. Sampling of the Knee Joint

Prepare one culture dish filled with 20 mL of PBS. Use ethanol 70% (v/v) to
disinfect and wet the rabbit’s fur all over the hind limbs, then cut the skin
around the ankle, pull the skin off the leg and the thigh, and roughly remove
the muscles with the scissors. Cut the femur and the tibia 2–3 cm away from
the knee, and then put the joint in the cell culture dish and transfer to a sterile
tissue culture hood (Fig. 5B).
Fig. 5. Animal sacrifice and cartilage collection. (A) Rabbit anesthesia using Minerve
apparatus and animal preparation before knee isolation. (B) Knee isolation; arrows show
dissection points. (C) Isolated femur; grid incision in the cartilage between the condylar
ridges; explants collected in the culture dish.
3.1.3. Cartilage Explant Preparation

All the following steps of explant preparation and culture are performed
under sterile conditions. Have ready two more culture dishes, one for tissue
dissection, and the other for collection of cartilage explants (see Note 2).
Prepare the DMEM/F12 for collection of isolated cartilage explants. Two
alternative supplements are used, depending on whether the explants are used
for proteoglycan or collagen degradation.
Fig. 6. Location of the sampling zone in the articular cartilage of the distal femur.
1. Proteoglycan degradation: 10% FBS DMEM/F12.

2. Collagen degradation: 0.1% BSA DMEM/F12.
Add 25 mL of the appropriate medium to one culture dish.
1. Remove the muscles around the joint, and cut the capsule and the ligaments to
separate the femur from the tibia. Transfer the femur into a clean culture dish and
cover with PBS. Repeat the operation for each knee.
2. With the no. 11 scalpel, incise a grid in the cartilage between the condylar ridges,
of squares of approx 1 mm per side (Figs. 5C and 6), reaching with the tip of the
blade the underlying hard tissue (see Note 3).
3. Gently shave the cartilage as close as possible to the underlying hard tissue and
place the explants into the culture dish containing the appropriate medium. Repeat
the operation until all the cartilage in the gridded zone has been collected.
4. Use the no. 23 scalpel to trim the largest explants already collected in the culture
dish (see Note 4).
3.2. Cartilage Explant Culture

3.2.1. Proteoglycan Degradation (Aggrecan; 35SO42–)
Two procedures for proteoglycan degradation assays are described, the first
for aggrecanase-, and the second for MMP-dependent release of radiolabeled
material. The different steps of the two procedures are outlined as a flow chart
in Fig. 7.
3.2.1.1. EXPLANT LABELING, COMMON TO BOTH PROCEDURES
1. Add 3.7 MBq (100 µCi) of 35SO42– to the culture dish containing the explants in
10% FBS DMEM/F12, usually before the weekend, and put in the tissue culture
incubator.
2. After 3 days in culture, remove the unincorporated radioactivity by six media
changes, each with 30 mL of medium, over 24 h using 0.1% BSA DMEM/F12.
Fig. 7. Assay procedures for aggrecanase- and matrix metalloproteinase (MMP)-

dependent proteoglycan degradation. APMA, p-aminophenylmercuric acetate; IL-1,
interleukin-1.
3.2.1.2. AGGRECANASE-DEPENDENT PROTEOGLYCAN DEGRADATION

1. Prepare treatments in 5-mL tubes. For each group, prepare 2.5 mL of 0.1% BSA
DMEM/F12, with or without treatments. The assay pattern is as follows:
a. One basal control group of medium plus vehicle at the same concentration as
in the treated groups.
b. One stimulated control group with IL-1β at a final concentration of 10 ng/mL
plus vehicle at the same concentration as in the treated groups.
c. Several treatment groups with IL-1β at a final concentration of 10 ng/mL plus
treatment. The concentration of the vehicle (usually DMSO) is made the same
for all treatment groups (see Note 5).
Each group is made up of eight wells. Distribute treatments from tubes to 96-
well plates, 250 µL/well (Fig. 8).
2. Transfer the fragments into the 96-well plate, at one fragment per well, and put in
the tissue culture incubator.
3. After 2 d of culture, stop the incubation and transfer each fragment into 500 µL of
papain solution/Pony Vial, then cover, and digest in the oven at 56°C for 16 h
(see Note 6). Then add 5 mL of LSC per vial and measure the radioactivity.
Fig. 8. 96-Well plate format for proteoglycan or collagen degradation.
4. Pipet an aliquot of 200 µL of culture media from each well into counting vials
and then add 5 mL of LSC. Measure the radioactivity, and multiply the dpm
result of counting by 1.25 to obtain the total radioactivity in the media per well.
5. Express proteoglycan degradation in each fragment as the percentage of released
radioactivity by the formula:
media radioactivity
degradation = × 100
media radioactivity + tissue radioactivity
3.2.1.3. MMP-DEPENDENT PROTEOGLYCAN DEGRADATION

1. After labeling and washing the explants, add 20 mL of 0.1% BSA, DMEM/F12
to the culture dish plus IL-1β at a final concentration of 10 ng/mL.
2. Incubate for 1 d and then wash one time with 25 mL of 0.1% BSA, DMEM/F12.
3. Prepare treatments in 5-mL tubes. For each group, prepare 2.5 mL of 0.1% BSA,
b. One stimulated control group with APMA at a final concentration of 500 µM
plus vehicle at the same concentration as in the treated groups.
c. Several treatment groups with APMA at a final concentration of 500 µM plus
treatment. The concentration of the vehicle (usually DMSO) is made the same
for all treatment groups (see Note 5).
Each group is made up of eight wells. Distribute treatments from tubes to 96-well
plates, 250 µL/well.
4. Transfer the fragments into the 96-well plate, at one fragment per well (Fig. 8),
and put in the tissue culture incubator.
5. After 1 d of culture, stop the incubation and transfer each fragment into 500 µL
of papain solution/Pony Vial, and then cover and digest in the oven at 56°C for
16 h (see Note 6). Then add 5 mL of LSC per vial and measure the radioactivity.
6. Pipet an aliquot of 200 µL of culture media from each well into counting vials
and then add 5 mL of LSC. Measure the radioactivity, and multiply the dpm
result of the media aliquot by 1.25 to obtain the total radioactivity in the media
per well.
Fig. 9. Procedure for collagen degradation assay. IL-1, interleukin-1.
7. Express proteoglycan degradation in each fragment as the percentage of released

radioactivity by the formula:
media radioactivity
media radioactivity + tissue radioactivity
3.2.2. Collagen Degradation (Collagen; OH-pro)

The different steps of the procedure are presented as a flow chart in Fig. 9.
3.2.2.1. TYPE II COLLAGEN DEGRADATION
1. Prepare treatments in 5-mL tubes. For each group, prepare 1.2 mL of 0.1% BSA,
b. One stimulated control group with IL-1β and plasmin, respectively, at a final
concentrations of 10 ng/mL and 0.1 U/mL plus vehicle at the same concentra-
tion as in the treated groups.
c. Several treatment groups with IL-1β and plasmin, at final concentrations, of
10 ng/mL and 0.1 U/mL respectively, plus treatment. The concentration of
the vehicle (usually DMSO) is the same for all treatment groups (see Note 5).
Each group is made up of eight wells. Distribute treatments from tubes to 96-well
plates, 120 µL/well.
2. Transfer the fragments into the 96-well plate, at two fragments per well (Fig. 8),
and put in the tissue culture incubator.
3. After 2 d of culture, stop the incubation and collect media and fragments for
hydrolysis.
a. For media, mix 100 µL of culture medium from each well with 100 µL of
37% HCl in a borosilicate vial.
Fig. 10. 96-well plate format for hydroproline (OH-pro) colorimetric assay.
b. For explants, put both fragments from each well into a borosilicated vial con-
taining 100 µL of deionized water, and then add 100 µL of 37% HCl.
Close the vials tightly with the appropriate caps and hydrolyze in the oven
at 110°C for 12 h.
3.2.2.2. ASSAY OF HYDROXYPROLINE: MODIFIED GRANT’S METHOD

1. Transfer 100 µL from each vial into a 96-well plate, and add 75 µL NaOH, 6 M to
take the sample close to neutrality. Mix energetically with an orbital plate shaker
for 5 min. at 1000 rpm (see Note 7).
2. Standard curve preparation: add 500 µL of deionized water to eight tubes num-
bered from 0 (blank) to 7. Dilute 100 µL of 5 mM OH-pro stock solution into
900 µL of deionized water, then transfer 500 µL of this solution into tube 1, and
mix thoroughly. Transfer 500 µL from tube 1 into tube 2 and mix thoroughly.
Repeat this procedure to tube 7. You obtain a 7-point standard curve plus a
blank with a concentration range from 250 to 3.91 µM by 1:2 dilutions. Prepare
a standard curve each time the assay is performed.
3. For the reaction setup, use a 1.2-mL 96-well deep plate; assay each standard
point in duplicate. Add 50 µL of standard or neutralized sample per well follow-
ing the pattern depicted in Fig. 10.
4. Add 185 µL of CT solution and mix gently on an orbital shaker for 5 min at
600 rpm.
5. Add 75 µL of pDAB solution and mix energetically on an orbital shaker for 2
min at 1000 rpm.
6. Cover the plate and incubate for 5 min at 70°C in a water thermostated bath.
7. Let the plate cool down at room temperature and transfer, using a multichannel
pipet, 250 µL from each well into a 96-well microtiter plate.
8. Read the plate at 540 nm within 15 min.
3.2.2.3. CALCULATING THE RESULTS

1. Plot OD540 vs concentration of standards. Read unknowns from the standard curve
(Fig. 11).
Fig. 11. Standard curve for hydroproline (OH-pro) assay. OD540 corrected from blank.
2. To obtain the total content of OH-pro per well, multiply each value from media
and fragments by the appropriate factor; see the following formulas:
VT × VH × VN
Fmedium = 3 2
= 0.42
V M × 10 × 10
VH × VN
Fexplant = 3 2
= 0.35
10 × 10
VT = treatment volume = 120 µL.
VH = hydrolyzed solution volume = 200 µL.
VN = volume of neutralized solution = 175 µL.
VM = volume of hydrolyzed medium = 100 µL.
3. Collagen degradation in each fragment is expressed as the percentage of released
OH-pro by the formula:
media OH-pro
media OH-pro + tissue OH-pro
3.3. Statistical Analysis

For all degradation protocols, namely, aggrecanase- and MMP-dependent
proteoglycan degradation and collagen degradation, the data are treated in the
same way.
1. Calculate mean and standard error of degradation for each group.
2. Validate the assay by verifying the stimulation of degradation by comparing the

mean of the basal control with that of the stimulated group without treatment. For
these two groups calculate the percentage of variation using the formula:
y –x
stimulated control
% variation = × 100
x
control
where x and y are, respectively, the degradation means of the control and stimu-
lated-without-treatment groups (see Note 8).
3. For treated groups, calculate the percentage of inhibition in each group using the
formula:
y –z
stimulated treated
% inhibition = × 100
y x
stimulated control
where z is the degradation mean of the treated group.

4. Compare basal control and stimulated-without-treatment group by unpaired t-test.
Compare stimulated-without-treatment and treated groups by analysis of variance
followed by a Dunnett’s multiple comparison test.
4. Notes
1. Solubilize APMA by fast stirring using a magnetic bar, or by sonication.
2. Always keep cartilage sample in PBS buffer or medium, thus avoiding tissue
drying.
3. In preliminary experiments (data not shown), it was observed that the cartilage
fragments from this region expressed a stronger response to degradative agents
such as retinoic acid and IL-1 than tissue fragments from femoral condyles, or
from femoral or humeral heads. Even if cartilage from different regions can be
used, it is strongly recommended not to mix tissue fragments collected from dif-
ferent locations in the same assay.
4. It is important that cartilage samples be homogenous in size and shape in order to
limit variability, especially for collagen degradation.
5. All groups are matched for concentration of vehicle (i.e., DMSO), generally not
more than 1% at final concentration.
6. During digestion of cartilage fragments, put a container filled with water in the
oven, thus reducing evaporation of papain solution.
7. For the assay of OH-pro, a precipitate may occur during the neutralization step,
which usually disappears during stirring and has no effect on the final result.
8. Typical values for degradation in all protocols are given below (data are mean
standard error of seven experiments):
a. Aggrecanase dependent proteoglycan degradation: results expressed as the
percentage of released radioactivity.
i. Basal control = 13.33 0.71
ii. Stimulated group without treatment = 44.00 1.82

iii. Percentage of stimulation = 235 19
b. MMP-dependent proteoglycan degradation: results expressed as the percent-
age of released radioactivity.
c. Type II collagen degradation: results expressed as the percentage of released
OH-pro.
References
1.
1 Goldring, M. B. (2000) The role of chondrocyte in osteoarthritis. Arthritis Rheum.
43, 1916–1926.
2.
2 Hascall, V. C. and Kimura, J. H. (1992) Proteoglycans: isolation and character-
ization. Methods Enzymol. 82, 769–800.
3.
3 Tortorella, M. D., Burn, T. C., Pratta, M. A., et al. (1999) Purification and cloning
of aggrecanase-1: a member of the ADAMTS family of proteins. Science 284,
1664–1666.
4.
4 Abbaszade, I, Liu, R. Q., Yang, F., et al. (1999) Cloning and characterization of
ADAMTS11, an aggrecanase from the ADAMTS family. J. Biol. Chem. 274,
23,443–23,450.
5.
5 Caterson, B., Flannery, C. R., Hughes, C. E., and Little, C. B. (2000) Mechanisms
involved in cartilage proteoglycan catabolism. Matrix Biol. 19, 333–344.
6.
6 Sabatini, M., Bardiot, A., Lesur, C., et al. (2002) Effects of agonists of peroxi-
some proliferator-activated receptor γ on proteoglycan degradation and matrix
metalloproteinase production in rat cartilage in vitro. Osteoarthritis Cartilage 10,
673–679.
7. Sorbera, L. A. and Castaner, J. (2000) Prinomastat. Drugs Fut. 25, 150–158.
8.
8 Sugimoto, K., Takahashi, M., Yamamoto, Y., Shimada, K., and Tanzawa, K.
(1999) Identification of aggrecanase activity in medium of cartilage culture.
J. Biochem. 126, 449–455.
9.
9 Cremer, M. A., Rosloniec, E. F., and Kang, H. A. (1998) The cartilage collagens;
a review of their structure, organization, and role in the pathogenesis of experi-
mental arthritis in animals and in human disease. J. Mol. Med. 76, 275–288.
10.
10 Billinghurst, R. C., Dahlberg, L., Iionescu, M., et al. (1997) Enhanced cleavage of
type II collagen by collagenases in osteoarthritic articular cartilage. J. Clin. Invest.
99, 1534–1545.
11.
11 Saito, S., Katoh, M., Masumoto, M., Matsumoto S.-I., and Masuho Y. (1997)
Collagen degradation induced by the combination of IL-1α and plasminogen in
rabbit articular cartilage explant culture. J. Biochem. 122, 49–54.
12. Kummer, J. A., Abbink, J. J., Deboer, J. P., et al. (1992) Analysis of intraarticular
fibrinolytic pathways in patients with inflammatory and non inflammatory joint
diseases. Arthritis Rheum. 36, 884–893.
13. Grant, R. A. (1964) Estimation of OH-proline by the autoanalyser. J. Clinical
Pathol. 17, 685–686.
Aggrecan Neoepitope Antibodies 237
17
Production of Antibodies Against
Degradative Neoepitopes in Aggrecan
John S. Mort and Peter J. Roughley
Summary
The use of synthetic peptides to generate rabbit polyclonal anticatabolic neoepitope anti-
bodies that can be used to study the presence of defined proteolytic cleavage sites in aggrecan
is described. Principles of peptide design and methods for preparation and characterization of
ovalbumin conjugates are presented along with approaches for the characterization and affinity
purification of the resulting antisera. Limitations associated with the use of antipeptide anti-
bodies to study authentic protein neoepitopes are discussed.
Key Words: Peptide synthesis; affinity purification; coupling; bifunctional reagent; ELISA;
SDS-PAGE; HPLC; peptide sequencing.
1. Introduction
Antibodies with the ability to recognize specific epitopes generated follow-
ing the cleavage of proteoglycans by different proteolytic enzymes (anti-
catabolic neoepitope antibodies) have played an important role in the charac-
terization of the proteases active within cartilage during normal development
(1) and degeneration of this tissue in arthritis (2). These antibodies appear to
bind the terminal residues of the epitope into a pocket in the antigen binding
site and are thus unable to recognize uncleaved proteoglycans (3,4).
For the production of antineoepitope antibodies, synthetic peptides repre-
senting the appropriate sequence upstream or downstream of the cleavage site
are prepared with an additional spacer sequence and a terminal cysteine residue
to allow coupling to a protein carrier using a bifunctional reagent (5). Although
other strategies can be used, we have had good success using peptide-ovalbu-
min conjugates coupled using the bifunctional reagent N-succinimidyl
bromoacetate (Fig. 1). Ovalbumin has several advantages as the carrier protein.
237
238 Mort and Roughley
Fig. 1. Schematic representation of the steps in the coupling of peptides to ovalbu-

min using the bifunctional reagent N-succinimidyl bromoacetate.
It is highly soluble, even following derivatization, and it runs as a discrete band

on sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE),
making it easy to characterize the conjugated products. It contains 20 lysine
residues for derivatization by a coupling reagent, no free sulfhydryl groups, and
its N-terminus is blocked by acetylation. Reaction with N-succinimidyl
bromoacetate results in a well-derivatized protein ready to bind free sulfhydryl
groups in a peptide with the introduction of a minimal amount of nonpeptide
material (a methylene group), thus decreasing the potential for antibody pro-
duction against physiologically irrelevant epitopes.
Two approaches have been used for neoepitope antibody generation. Mouse
monoclonal antibodies have the advantage of being unique immunoglobulins
that can, in principle, be produced in unlimited quantities. Effective monoclonal
antibodies have been produced to aggrecan cleavage epitopes (6,7). However,
considerable time and effort must be invested to generate the required hybri-
doma clones. In contrast, rabbit polyclonal antibodies are much easier to pro-
duce, but the resulting sera of necessity contain a heterogeneous population of
immunoglobulins. This can result in polyclonal antisera being less specific
for the neoepitope than monoclonal antibodies. Affinity purification, using the
immobilized immunizing peptide, removes antibodies to ovalbumin and other
serum components contributing to high background reactivity on Western blot-
ting and immunohistochemistry.
In this chapter, principles of peptide design and methods for preparation and
characterization of ovalbumin conjugates are outlined, along with approaches
for the production, characterization, and affinity purification of the resulting
rabbit immunoglobulins. In addition, some possible pitfalls associated with the
use of these antibodies are discussed.
2. Materials
1. Peptides can be obtained either commercially or by synthesizing in-house. Puri-
fication by high-performance liquid chromatography (HPLC) and characteriza-
tion by mass spectrometry or peptide sequencing is required.
2. DL-Cysteine (Sigma, cat. no. C 4022).
3. Ellman’s reagent (8): 5,5'-dithiobis (2-dinitrobenzoic acid) (Sigma, cat. no. D
8130). Make a stock solution of 4 mg/mL in 0.1 M potassium phosphate, pH 7.0,
containing 0.05% sodium azide, and (store at 4°C).
4. Ellman’s reaction buffer: 0.1 M potassium phosphate, pH 8.0, containing 1 mM
EDTA and 0.05% sodium azide. (Store at room temperature.)
5. Ellman’s reaction mix: 4 mL Ellman’s reaction buffer and 0.12 mL Ellman’s
reagent solution. Make up to 14 mL with water.
6. Dithiothreitol (Sigma, cat. no. D 9163).
7. Ovalbumin (Sigma, cat. no. A 5503; albumin from chicken egg white), stored
at 4°C.
8. Coupling reagent: the bifunctional reagent N-succinimidyl bromoacetate is avail-
able from Sigma-Aldrich (cat. no. B8271, bromoacetic acid N-hydroxy-
succinimide ester) or can be synthesized with little difficulty (9). It is extremely
important that this compound be stored under dry conditions. Before use, the
bifunctional reagent is dissolved in dimethylformamide. This should be stored
over molecular sieves (3 Å) to prevent decomposition resulting in the formation
of ammonia, which would react with the coupling reagent.
9. Coupling buffer and column buffer: 0.1 M potassium phosphate, pH 7.5, contain-
ing 1 mM EDTA. (Adjust with 10% potassium hydroxide.) Filter and de-gas.
10. Phosphate-buffered saline (PBS): 6 mM Na2HPO4, 4 mM KH2PO4, 145 mM NaCl,
pH 7.2.
11. SDS-PAGE equipment.
12. Electrophoretic transfer equipment.
13. PVDF (polyvinylidinedifluoride) and nitrocellulose membranes (Bio-Rad).
14. Sephadex G10 and G25 columns, (Amersham Biosciences).
15. New Zealand white rabbits (female, 2.5–3.0 kg).
16. Freund’s complete and incomplete adjuvants (Difco).
17. Immunlon 2 flat-bottomed 96-well ELISA plates.
18. PBS-T: PBS containing 0.05% Tween-20.
19. Blocking buffer: PBS-T containing 1% bovine serum albumin.
20. Alkaline phosphate-conjugated goat anti-rabbit immunoglobulins (Sigma, cat.
no. A-3812).
21. Alkaline phosphatase substrate (p-nitrophenylphosphate: Sigma Diagnostics, cat.

no. 104-105). Dissolve 1 tablet in 10 mL diethanolamine buffer (9.6% v/v
diethanolamine in 0.26 mM MgCl2, pH 9.8).
22. TBS-T: Tris-buffered saline (10 mM Tris-HCl, pH 8.0, 150 mM NaCl) contain-
ing 0.05% Tween-20.
23. Alkaline phosphatase blot development solution: add 66 µL of nitroblue tetrazo-
lium (50 mg/mL in dimethylformamide) and 33 µL of 5-bromo-4-chloro-3-
indolyl phosphate (50 mg/mL in dimethylformamide) in 10 mL of alkaline
phosphatase buffer (100 mM Tris-HCl, pH 9.5, 100 mM NaCl, 5 mM MgCl2).
24. Sulfolink coupling gel (Pierce)
25. Affinity column equibration buffer: 50 mM Tris-HCl, pH 8.5, containing 5 mM
EDTA.
3. Methods
3.1. Peptide Design
As a general rule, peptides are prepared consisting of five to seven amino
acid residues representing the sequence immediately upstream, or downstream,
of the target cleavage site, as appropriate. Two glycine residues are then rou-
tinely added as a spacer to facilitate accessibility of the epitope, followed by a
cysteine residue to allow coupling through the sulfhydryl group on the side
chain. Hydrophobic residues may predominate in the target sequence and this
would result in peptides with low water solubility. In such cases the addition of
arginine and lysine residues to the spacer sequence has proved helpful. GKG,
AKKG, and GKARAKG linkers have been used successfully (10) (Fig. 2). A
basic linker region may also help solubilize acidic peptides for reverse-phase
HPLC purification using trifluoroacetic acid/acetonitrile (see Note 1).
To characterize fully the specificity of the antipeptide antisera, additional
peptides representing a one-residue truncation and a one-residue extension of
the neoepitope sequence are synthesized together with the immunizing peptide
for use in a competitive ELISA.
Success of peptide coupling depends on the status of the cysteine sulfhydryl
group. Often peptides are partially, or occasionally completely, oxidized. The
sulfhydryl content of the peptide can be determined. In some cases a reduced
peptide can be regenerated from the oxidized form by treatment with a reduc-
ing agent such as dithiothreitol.
3.2. Testing for Free Sulfhydryl Content of Peptides
3.2.1. Standard Curve
1. Prepare a fresh 4 mM solution of cysteine in water and make a working solution
of 0.4 mM in water.
2. Prepare 0.3 mL serial dilutions of cysteine in water from 0 to 0.4 mM in incre-
ments of 40 µM.
Fig. 2. Schematic representation of the aggrecan core protein and peptides designed
for production of anti-neoepitopes for well-characterized metalloproteinase cleavage
sites. The aggrecan core protein is depicted as an N-terminal globular region (G1) sepa-
rated from a second globular region (G2) by an interglobular domain (IGD). The IGD is
susceptible to cleavage by many proteolytic enzymes. Shown are the aggrecanase (A1)
and matrix metalloproteinase (M) cleavage sites. Although some keratan sulfate substi-
tution is present in the G1 and IGD, the predominant sites of attachment are in a region
(KS) immediately following the G2 region. This is followed by two extended regions of
chondroitin sulfate substitution (CS1 and CS2). Four aggrecanase cleavage sites are
located in the CS2 domain, of which only one is shown for convenience (A2). The
aggrecan core protein is terminated with a third globular region (G3). Peptides designed
for production of antibodies to detect the products of aggrecanase cleavage in the IGD
(A1) followed the general principle of including the six residues bordering the cleavage
site (bold font), followed by the standard linker region (normal font). For the most
N-terminal cleavage site in the CS2 region (A2), redundant peptides were used in order
to produce antisera that would recognize both human (first alternative) and bovine (sec-
ond alternative) aggrecan cleavage products. In both cases a single peptide synthesis
was carried out, and at cycle six an equimolar mixture of appropriately blocked serine
and glycine or glycine and aspartic acid was used. No effort was made to purify the
resulting peptides after cleavage from the resin. The new N-terminus generated by
matrix metalloproteinases in the IGD (M) mostly consists of hydrophobic residues. In
this case, positively charged amino acids were added to the linker region to assist solu-
bility of the peptide.
3. To each tube add 0.7 mL of Ellman’s reaction mix.

4. Read absorbance at 412 nm and plot standard curve.
3.2.2. Sulfhydryl Estimation

1. Accurately weigh 1–2 mg of peptide and dissolve in water to a final concentra-
tion of 3 mM (see Note 2).
2. Prepare a 1:10 dilution and assay as in Subheading 3.2.1.
3. Estimate free sulfhydryl content from standard curve.
3.3. Reduction of Sulfhydryl Groups on Peptides

3.3.1. Gel Filtration Column
1. Swell approx 10 g of Sephadex G10 in 1% acetic acid overnight and pack a 30 ×
1 cm column. (Bio-Rad Econo columns are good.)
2. Equilibrate the column with 1% acetic acid at a flow rate of 1 mL/min.
3.3.2. Peptide Reduction

1. Dissolve the peptide at 2 mg/mL in 25 mM Tris-HCl, pH 7.5.
2. Add solid dithiothreitol to give a final concentration of 10 mM.
3. Purge solution for 2–3 min with nitrogen and leave at room temperature for at
least 1 h.
4. Apply the reduced peptide sample to the Sephadex G10 column (see Subhead-
ing 3.3.1.), and elute with 1% acetic acid, collecting 1-mL fractions.
5. Assay the fractions for free sulfhydryl as in Subheading 3.2.2.
6. Pool the first peak and freeze-dry the reduced peptide.
3.4. Ovalbumin Activation and Peptide Coupling

3.4.1. Gel Filtration Column
1. Swell Sephadex G25 (medium) in water for at least 4 h. One gram of resin gives
4–6 mL of gel.
2. At room temperature, pack a 30 × 1-cm column. (Bio-Rad Econo columns are
good.) This requires about 25 mL of resin. Equilibrate the column with coupling
buffer using a flow rate of 2–3 mL/min.
3. After use the column should be washed with at least 2 bed vol of coupling buffer
containing 0.05% sodium azide for storage. For future use, at least 2 column vol of
coupling buffer should be passed through the column to remove the sodium azide.
3.4.2. Preparation of Activated Ovalbumin

1. Prepare an ice bath for use on a magnetic stirrer.
2. Weigh out 50 mg of ovalbumin into an 8-mL Wheaton vial, and add 2.25 mL of
coupling buffer and a small magnetic stirring bar. Place the Wheaton vial in the
ice bath and stir slowly to dissolve the ovalbumin.
3. Cool 1 mL dimethylformamide in a borosilicate tube. Make at least 0.2 mL of a 65
mg/mL solution of N-succinimidyl bromoacetate in the cold dimethylformamide.
4. Over the course of about 1 min add, dropwise, 0.2 mL of the N-succinimidyl
bromoacetate solution to the dissolved ovalbumin with moderate stirring. Allow
the reaction to incubate in the ice bath for 5 min. Lift the vial out of the ice bath
and allow the vial to warm to room temperature over the course of 25 min.
5. Apply the solution of activated ovalbumin to the Sephadex G25 column and col-
lect 25 1.5-mL fractions at a flow rate of about 1.5-mL/min.
6. Read the absorbance at 280 nm of 1:10 or 1:20 dilutions in water of the column
fractions. The profile will consist of two peaks. The first peak is the derivatized
ovalbumin.
7. Pool the major fractions of the first peak (usually two or three), making sure not
to include any trace of the second peak.
8. The activated ovalbumin is best used immediately, but it can be stored for up to
1 wk at 4°C in the dark.
3.4.3. Coupling of Peptides to Activated Ovalbumin (see Note 2)

1. Dissolve 3 mg of the peptide for coupling in 0.4 mL of water (see Note 3) in a 4-mL
Wheaton vial.
2. As a control, 0.4 mL of 6 mM cysteine (0.73 mg/mL) in water is used.
3. Add 0.4 mL of the activated ovalbumin to each peptide or the cysteine solution.
Immediately purge the vials with nitrogen for about 30 s and allow the coupling to
proceed at room temperature for 2 h on a rocking platform with the vials lying on
their sides. Subsequently leave the vials on the rocking platform overnight at 4°C.
4. Then add 2-mercaptoethanol (1 µL) and leave the solution for 1 h to block
unreacted sites on the ovalbumin.
5. Estimate the efficiency of coupling by SDS-PAGE (see Subheading 3.5.)
6. Dialyze conjugate solutions, against three changes of PBS-azide and keep at 4°C
for immediate use or at –20°C for long-term storage.
3.5. Estimation of Coupling Efficiency

Before using conjugates for immunization, it is advisable to check that good
peptide substitution has been obtained. Decreased mobility on SDS-PAGE rela-
tive to an ovalbumin-cysteine control conjugate represents a simple qualitative
demonstration that substantial coupling has occurred (Fig. 3).
Although a substantial retardation on SDS-PAGE is a good indication that
coupling has occurred, some conjugates that exhibit minimal retardation have
produced good antibodies. Before writing off apparently failed conjugates, it can
be worth checking for bound peptide by N-terminal sequencing of the conjugate.
To ensure complete removal of free peptide prior to sequencing, conjugates are
separated by SDS-PAGE, blotted to a PVDF membrane, and stained briefly
with Coomassie blue. The target bands are excised and subjected to N-terminal
sequencing. Since the N-terminus of ovalbumin is blocked (acetylated), only the
sequence of the peptide should be observed. A rough estimate of the amount of
peptide bound can be made from the yield of several cycles. In the case of pep-
tides coupled through an N-terminal cysteine residue, the first cycle will be blank,
but subsequent cycles give a good representation of the expected signal.
3.5.1. Mobility Shift on SDS-PAGE

1. Prepare 10% SDS-PAGE gels (11).
2. Run 5 µL of peptide and cysteine conjugate solutions.
3. Stain with Coomassie blue (12). Compare mobility of peptide conjugates relative
to that to of the cysteine-ovalbumin control.
Fig. 3. Evaluation of coupling efficiency by SDS-PAGE. Activated ovalbumin was

coupled to cysteine (C, control) and to four different peptides (1–4). A clear shift in
mobility was seen for peptides 1 and 2 indicating extensive derivatization. In the case
of peptides 3 and 4 little difference in mobility was seen relative to the cysteine con-
trol, although a slight difference in the appearance of the bands is discernable. In such
instances more detailed analysis of the putative conjugates is in order, for example by
N-terminal sequencing, to determine whether any peptide is coupled. In these particu-
lar examples enough peptide was in fact present to yield useful antisera. However, in
most cases where the migration pattern of the reacted ovalbumin closely resembles the
cysteine control, failure of coupling should be suspected.
3.6. Antibody Production

Although immunization of one rabbit per conjugate can be successful, it is
often wise to use several animals to increase the odds of producing an antibody
with the required properties.
3.6.1. Immunization
1. Obtain a preimmune bleed (5 mL) from each rabbit to be used as control serum
later.
2. In a 1.5-mL Eppendorf tube, add peptide conjugate and enough PBS to result in
0.75 mg of peptide conjugate in 0.375 mL. Add 0.375 mL of Freund’s complete
adjuvant. Cap the tube and agitate violently on a vortex mixer to produce a stiff
emulsion. It should be possible to draw enough emulsion into a syringe to inject
the rabbit intramuscularly on each flank with 0.25 mL of emulsion (total 0.5 mg
of antigen).
3. After 2 wk, boost the rabbits with a similar mixture produced using Freund’s
incomplete adjuvant.
4. After a further 10 d, take a test bleed and assess the antigenic response. If this is
judged to be adequate, bleed out the animal. (Up to 150 mL of blood should
obtainable.) The use of 50 mL non-anticoagulant centrifuge tubes is recom-
mended. Allow the blood to clot and store at 4°C overnight to allow the clot to
contract. Remove the serum and centrifuge at low speed to remove any contami-
nating red blood cells. Store serum frozen in aliquots.
5. Alternatively, if the antigenic response is not adequate administer, a second boost
with emulsion containing Freund’s incomplete adjuvant and bleed out the animal
after a further 10 d.
3.7. Antibody Characterization

The antigenic response of the different rabbits can be determined by direct
enzyme-linked immunosorbent assay (ELISA) using adsorbed peptide.
Neoepitope specificity can be demonstrated by competitive ELISA, using pep-
tides representing truncated or extended versions of the true epitope. Lack of
reaction to a peptide spanning the cleavage site is also a good demonstration of
specificity (13). However, it is critical that reactivity of the antibody with the
authentic protein cleavage site epitope also be demonstrated, as the same anti-
bodies may react well with the immunizing peptide but not with the native
protein (see Subheading 3.9.).
3.7.1. Direct ELISA

1. Coat ELISA plate wells with 50 ng of peptide in 50 µL of 0.01 M sodium bicar-
bonate, pH 9.6, overnight at 4°C.
2. Wash wells with PBS-T.
3. Block wells for 1 h at room temperature with 100 µL of blocking buffer.
4. Add serial antiserum dilutions (50 µL) from 1:10 to 1:100,000 in blocking buffer
and incubate at 37°C for 60 min.
6. Add 50 µL of alkaline phosphatase-conjugated goat anti-rabbit IgG (1:1000 dilu-
tion) and incubate at 37°C for 60 min.
8. Add 50 µL of alkaline phosphatase substrate and allow color to develop for 15
min at 37°C.
9. Read absorbance at 405 nm on a plate reader.
3.7.2. Competitive ELISA to Demonstrate Neoepitope Specificity

1. Coat plates and block wells as in steps 1–3 in Subheading 3.7.1.
2. Use a constant dilution of antiserum (previously demonstrated to give 50% of
maximum binding) and make serial dilutions of competing peptide ranging
from 25 to 150,000 ng/mL in this antibody dilution. Incubate for 1 h at room
temperature.
3. Add the antibody/peptide complexes to the wells.
4. Develop as in steps 5–9 in Subheading 3.7.1.
5. Plot results and calculate the mean inhiting concentration (IC50) value for inhibi-
tion (Fig. 4).
Fig. 4. Competitive ELISA analysis to demonstrate antibody selectivity. An anti-

body to the matrix metalloproteinase cleavage epitope ...DIPEN was produced as
described in this chapter. Its specificity was investigated by competitive ELISA using
CGGVDIPEN as the bound peptide. When used as a competitive antigen, this
sequence gave a median inhibitory concentration (IC50) of 0.03 µM. In contrast, exten-
sion of the epitope by one residue to ..DIPENF resulted in a greater than three orders
of magnitude increase in IC50, indicating the specificity of the antibody for the termi-
nal epitope. In contrast, when the homologous cleavage sequence for bovine aggrecan
..DIPES was used, the IC50 was 3 µM, indicating that some flexibility can be tolerated,
albeit at decreased reactivity.
3.7.3. Western Blotting of Authentic Protein Cleavage Product

1. Run samples of tissue extract or in vitro digest on SDS-PAGE as duplicate sets.
2. Electroblot onto nitrocellulose or PVDF membranes (14).
3. Block the membrane in blocking solution for 1 h.
4. Wash membrane in TBS-T.
5. Cut membrane into separate sample sets.
6. Dilute anti-neoepitope antibody in blocking solution (1:500 is usually a good
dilution) and incubate with one membrane for 1 h at room temperature.
7. To determine antibody specificity, repeat step 6 on the membrane from the
duplicate sample set using a 1:500 dilution of antibody containing 1 mg of
epitope peptide for every mL of serum (2 µg/mL peptide in a 1:500 serum dilu-
tion). This mixture should be left for 1 h before use.
8. Wash membranes in TBS-T.
9. Incubate membranes in alkaline phosphatase-conjugated goat anti-rabbit immu-
noglobulin 1:7500 in blocking solution for 30 min at room temperature.
10. Wash the membranes in TBS-T.
11. Add alkaline phosphatase blot development solution and allow bands to develop.
12. Wash blots with water.
3.8. Affinity Purification

To reduce background staining, it is often advisable to purify the antineo-
epitope antibodies by affinity chromatography on a resin containing the pep-
tide used for immunization.
3.8.1. Affinity Column Preparation

1. Allow the Sulfolink gel to equilibrate to room temperature.
2. Pack a 10-mL disposable column with 1.5 mL of gel slurry.
3. Equilibrate the column with 10 mL of affinity column equilibration buffer and
allow excess buffer to drain. Close the end of the column.
4. Add 1.5 mL of 2 mg/mL peptide in water to the gel. Cap the column and mix by
inverting frequently over the course of 15 min. Leave at room temperature for a
further 30 min.
5. Drain and wash the column with 5 mL of equilibration buffer.
6. To block unreacted sites on the resin, add 1.5 mL of 50 mM cysteine in equilibra-
tion buffer and mix as above.
7. Drain and wash the column with equilibration buffer.
8. After use, the column should be washed with 20 mL of 1 M NaCl and then stored
in water containing 0.05% sodium azide at 4°C.
3.8.2. Affinity Purification

1. Equilibrate appropriate Sulfolink column with affinity column equilibration
buffer.
2. Centrifuge serum (10 mL) at medium speed to remove any particulates and apply
clarified serum to column.
3. Collect the flowthrough of the serum. (This may still contain useful antibody.)
4. Wash column with 25 mL of PBS-azide.
5. Apply 15 mL of 0.1 M glycine, pH 2.8, and collect 1 mL fractions into tubes
containing 50 µL of 1 M Tris-HCl, pH 9.5, to neutralize the eluted immunoglo-
bulin.
6. Read the absorbance of the fractions at 280 nm.
7. Dialyze the peak fractions against PBS overnight at 4°C.
8. Wash the column for storage as described in Subheading 3.7.1.
3.9. Pitfalls
The generation of antipeptide anti-neoepitope antibodies has several poten-
tial limitations that may interfere with their use for the analysis of physiologi-
cally and pathologically relevant protein products.
1. The epitope may not be readily accessible to the antibody owing to steric block-
ing by an adjacent protein substituent such as a glycosaminoglycan chain of
aggrecan (Fig. 2). This can be overcome by chondroitinase treatment.
2. The amino acids within the epitope may themselves be amenable to post-transla-
tional modification. Such a scenario occurs within the carboxy-terminal
neoepitope generated by aggrecanases in the IGD. Here asparagine and threonine
residues in the sequence ...NITEGE may act as sites for N-linked and O-linked
glycosylation, respectively. Glycosylation of these residues can decrease
neoepitope antibody affinity for the authentic protein product (10).
3. These limitations can make even semiquantitation of neoepitope abundance by
visualization of antibody reaction products misleading.
4. Notes
1. An additional strategy is to substitute β-alanine for the glycine residues in the
spacer region. This allows the peptide content of the final conjugate to be deter-
mined by amino acid analysis.
2. It is useful to include a peptide that from previous experience is known to couple
well, as a positive control.
3. Given the diversity of possible amino acid sequences, solubility problems can be
encountered with some peptides. In our experience, most peptides dissolve in
water. Acidic peptides that have not been provided with a basic linker region can
be rendered soluble by the addition of an organic base such as N-ethylmorpholine.
We have also had success with dissolving peptides in dimethylformamide or dim-
ethylsulfoxide so that the final concentration of organic solvent in the coupling
reaction is no more than 20 or 50%, respectively.
Acknowledgments
This work was supported by the Shriners of North America, the Canadian
Institutes of Health Research, and the Arthritis Society of Canada.
References
1.
1 Lee, E. R., Lamplugh, L., Leblond, C. P., Mordier, S., Magny, M.-C., and Mort,
J. S. (1998) Immunolocalization of the cleavage of the aggrecan core protein at
the Asn341-Phe342 bond, as an indicator of the location of the metalloproteinases
active in the lysis of the rat growth plate. Anat. Rec. 252, 117–132.
2.
2 Lark, M. W., Bayne, E. K., Flanagan, J., et al. (1997) Aggrecan degradation in
human cartilage. Evidence for both matrix metalloproteinase and aggrecanase
activity in normal, osteoarthritic, and rheumatoid joints. J. Clin. Invest. 100,
93–106.
3.
3 Mort, J. S., Flannery, C. R., Makkerh, J., Krupa, J. C., and Lee, E. R. (2003) The
use of anti-neoepitope antibodies for the analysis of degradative events in carti-
lage and the molecular basis for neoepitope specificity. Biochem. Soc. Symp. 70,
107–114.
4. Mort, J. S. and Buttle, D. J. (1999) The use of cleavage site specific antibodies to
delineate protein processing and breakdown pathways. J. Clin. Pathol. Mol.
Pathol. 52, 11–18.
5.
5 Hughes, C. E., Caterson, B., White, R. J., Roughley, P. J., and Mort, J. S. (1992)
Monoclonal antibodies recognizing protease-generated neoepitopes from carti-
lage proteoglycan degradation: application to studies of human link protein
cleaved by stromelysin. J. Biol. Chem. 267, 16,011–16,014.
6. Hughes, C. E., Caterson, B., Fosang, A. J., Roughley, P. J., and Mort, J. S. (1995)
Monoclonal antibodies that specifically recognize neoepitope sequences gener-
ated by ‘aggrecanase’ and matrix metalloproteinase cleavage of aggrecan: appli-
cation to catabolism in situ and in vitro. Biochem. J. 305, 799–804.
7.
7 Fosang, A. J., Last, K., Gardiner, P., Jackson, D. C., and Brown, L. (1995) Devel-
opment of a cleavage-site-specific monoclonal antibody for detecting
metalloproteinase-derived aggrecan fragments: detection of fragments in human
synovial fluid. Biochem. J. 310, 337–343.
8.
8 Ellman, G. L. (1959) Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82, 70–77.
9.
9 Bernatowicz, M. S. and Matsueda, G. R. (1986) Preparation of peptide-protein
immunogens using N-succinimidyl bromoacetate as a heterobifunctional
crosslinking reagent. Anal. Biochem. 155, 95–102.
10.
10 Sztrolovics, R., White, R. J., Roughley, P. J., and Mort, J. S. (2002) The mecha-
nism of aggrecan release from cartilage differs with tissue origin and the agent
used to stimulate catabolism. Biochem. J. 362, 465–472.
11. Gallagher, S. R. (1995) One-dimensional SDS gel electrophoresis of proteins, in
Current Protocols in Protein Science (Coligan, J. E., Dunn, B. M., Ploegh, H. L.,
Speicher, D. W., and Wingfield, P. T., eds.), John Wiley & Sons, New York, NY,
pp. 10.1.1–10.1.34.
12. DeSilva, T. M., Ursitti, J. A., and Speicher, D. W. (1995) Protein detection in gels
using fixation, in Current Protocols in Protein Science (Coligan, J. E., Dunn, B.
M., Ploegh, H. L., Speicher, D. W., and Wingfield, P. T., eds.), John Wiley &
Sons, New York, NY, pp. 10.5.1–10.5.12.
13.
13 Mort, J. S., Magny, M.-C., and Lee, E. R. (1998) Cathepsin B: an alternative
protease for the generation of an aggrecan “metalloproteinase” cleavage
neoepitope. Biochem. J. 335, 491–494.
14. Ursitti, J. A., Mozdzanowsli, J., and Speicher, D. W. (1995) Electroblotting from
polyacrylamide gels, in Current Protocols in Protein Science (Coligan, J. E.,
Dunn, B. M., Ploegh, H. L., Speicher, D. W., and Wingfield, P. T., eds.), John
Wiley & Sons, New York, NY, pp. 10.7.1–10.7.14.
Immunoassays for Types II and IX Collagens 251
18
Immunoassays for Collagens
in Chondrocyte and Cartilage Explant Cultures
R. Clark Billinghurst, Fackson Mwale,

Anthony Hollander, Mirela Ionescu, and A. Robin Poole
Summary
Quantitative immunoassays have been developed to measure the content, degradation, and
synthesis of types II and IX collagens in hyaline cartilages. Some of these assays and their
applications are described in this chapter. These and other assays are commercially available.
The applications of these assays are discussed with examples from recent publications.
Key Words: Cartilage; type II collagen; type IX collagen; immunoassay; degradation; col-
lagenase.
1. Introduction
Type II collagen is the major structural component of articular cartilage
and provides this tissue with its tensile strength. It exists as fibrils of
crosslinked helical molecules composed of three identical α chains of approx
1000 amino acids each, with nonhelical extensions called telopeptides at both
ends of each chain. During turnover in health and osteoarthritis, after these
fibrils are depolymerized they are degraded (at neutral pH) by the proteolytic
attack of enzymes called mammalian collagenases at a specific locus within
the native triple-helical structure of each collagen molecule. These collagena-
ses belong to the matrix metalloproteinase (MMP) enzyme family and include
MMP-1 (collagenase-1 or fibroblast/interstitial collagenase), MMP-8 (colla-
genase-2 or neutrophil collagenase), and MMP-13 (collagenase-3). All three,
in addition to a membrane-type MMP (MT-MMP) designated MT1-MMP
(MMP-14), cleave the type II collagen molecules to yield characteristic 3/4
(TCA) and 1/4 (TCB) fragments.
251
252 Billinghurst et al.
The newly created carboxy (C) and amino (N) termini of the 3/4 and 1/4
fragments, respectively, represent neoepitopes that allow the specific identifi-
cation of collagenase-cleaved triple helical type II collagen molecules. Cleav-
age site neoepitopes have proved valuable in defining MMP activity in
aggrecan degradation (see also Chap 17). This chapter describes the proce-
dures used to develop an antibody and immunoassay for the detection and
quantitation of type II collagen bearing the C-terminal neoepitope of the 3/4
α-chain fragment created by cleavage by collagenases. This will include the
production and characterization of C1, 2C (formerly called COL2-3/4Cshort),
a polyclonal antibody that recognizes collagenase-cleaved fragments of type I
and II collagens (1). A monoclonal antibody, C2C (formerly called COL2-
3/4Clong mono) has also been developed to identify specifically collagenase-gen-
erated fragments of type II collagen. Immunoassays utilizing these antibodies
are commercially available (www.ibex.ca; IBEX Technologies, Montreal,
Quebec, Canada).
After helical cleavage, there is spontaneous denaturation (unwinding) of the
α-chains, which are susceptible to further degradation by gelatinolytic protein-
ases, such as MMP-2 (gelatinase A) and MMP-9 (gelatinase B). A specific
assay has been developed that recognizes a linear amino acid sequence from the
type II collagen α-chain (CB11B), which is “hidden” within the intact helical
structure but becomes exposed upon cleavage and denaturation of the triple
helical collagen molecule (2). The monoclonal antibody COL2-3/4m recognizes
this hidden epitope in denatured type II collagen and not the native molecule.
Sequential extraction of cartilage using α-chymotrypsin and proteinase K, com-
bined with assay of the extracts by an inhibition enzyme-linked immunosorbent
assay (ELISA) using the antibody COL2-3/4m, is a reliable method for assess-
ing the proportion of denatured type II collagen in cartilage and total type II
collagen content. The COL2-3/4m assay for type II collagen content and dena-
turation is also described in this chapter. An immunoassay is also available
(CPII; IBEX) for quantitating the levels of the C-terminal propeptide of type II
collagen that is cleaved from the procollagen molecule with fibrillogenesis and
is thus an indicator of type II collagen synthesis. This ELISA assay was devel-
oped from the radioimmunoassay we described previously (3).
The fibrillar organization of the extracellular matrix of articular cartilage
involves the assembly of collagen fibrils that contain not only type II collagen
but also types IX and XI collagens. Type IX collagen is a heterotrimer com-
posed of three genetically distinct chains designated α1(IX), α2(IX), and
α3(IX), forming three triple-helical domains (COL1, COL2, and COL3) alter-
nating with noncollagenous domains (NC1, NC2, NC3, and NC4). Type IX
molecules are organized in a periodic manner on the fibril surface. The α1-
chain has an N-terminal extension whereby the NC4 domain of the α1 (IX)
chain extends from the fibril surface into the perifibrillar space. Type IX col-
lagen is covalently crosslinked to type II collagen in an antiparallel orientation
and may also be crosslinked to other type IX collagen molecules. The amino-
terminal NC4 domain is very basic. Some forms of type IX collagen may lack
the NC4 domain because of the use of an alternative promoter, and the expres-
sion of this variant is thought to be tissue-specific and developmentally regu-
lated. This chapter describes the procedures used to develop four ELISA
immunoassays utilizing antibodies to the NC4 and COL2 domains of the
α1(IX) chain and to denatured type II collagen (COL2-3/4m) and collagenase
cleaved (C1, 2C) types I and II collagens (4). This includes the validation and
experimental use of the immunoassays.
2. Materials
2.1. Polyclonal Antibody Production
2.1.1. Preparation of the Immunogens
1. 0.1 M Phosphate buffer: 5.28 g/L KH2PO4, 16.4 g/L Na2HPO4 · 7H2O, 0.372 g/L
ethylenediaminetetraacetic acid (EDTA)-Na2, 0.2 g/L NaN3, pH 7.0.
2. Ovalbumin (OVA): 25 mg of grade V chicken egg albumin/mL of 0.1 M phos-
phate buffer.
3. Coupling reagent: 56.5 mg N-hydroxy-succinimidylbromoacetate/mL of chilled
N,N-dimethylformamide (Sigma).
4. Azide-free phosphate-buffered saline (PBS): 1.61 g/L Na2HPO4 · 7H2O, 8.5 g/L
NaCl, 0.55 g/L KH2PO4.
5. Sephadex G-25 column (Amersham; 27 × 2.2 cm) with 103 mL bed volume.
6. Synthetic peptide: 5 mg/100 µL of 0.1 M phosphate buffer.
7. Ultraviolet (UV) spectrophotometer and quartz cuvets.
2.1.2. Immunization
1. Peptide-OVA conjugate prepared in Subheading 3.1.1.
2. Azide-free PBS.
3. 5-mL Glass syringe and 20-gage needle.
4. Complete and incomplete Freund’s adjuvant (Difco).
5. Two female New Zealand white rabbits weighing 2.5–3.0 kg each.
2.1.3. Antibody Purification
1. Saturated ammonium sulfate solution: 761 g/L (NH4)2SO4, pH 7.0.
2. PBS: azide-free PBS + 0.5 g/L NaN3.
3. 0.8 M Sodium acetate buffer: 65.6 g/L C2H3O2Na, 4.0 g/L NaN3, pH 4.0.
4. 2% (w/w) Pepsin: 20 mg 2X crystallized pepsin A (Worthington Biochemi-
cal)/g IgG.
NaN3, pH 8.0.
6. Econo-Pac 10DG columns packed with Bio-Gel P-6 desalting gel (Bio-Rad).
7. HiTrap affinity column with 5 mL Protein A Sepharose High Performance gel

(Amersham).
8. Tris buffers: 121.1 g/L for 1 M, 12.1 g/L for 100 mM, and 1.21 g/L for 10 mM of
Tris(hydroxymethyl)aminomethane, pH 8.0.
9. 100 mM Glycine-HCl: 7.51 g/L, pH 3.0.
10. SulfoLink coupling gel and column (Pierce).
11. Coupling buffer: 50 mM Tris-HCl (6.06 g/L ), 5 mM EDTA-Na2 (1.86 g/L), pH 8.5.
12. Blocking reagent: 50 mM L-cysteine (6.1 g/L) in coupling buffer.
13. Washing solution: 58.44 g/L NaCl (1 M).
14. Dialysis tubing (molecular weight cutoff: 12–14,000).
15. UV spectrophotometer and quartz cuvets.
2.1.4. Antibody Characterization

1. 0.1 M Carbonate buffer: add 200 mL of 0.1 M Na2CO3 (10.6 g/L) to 0.1 M
NaHCO3 (8.4 g/L) until pH = 9.2.
2. PBS-Tween: PBS + 0.1% (v/v) polyoxyethylenesorbitan monolaurate (Tween-20).
3. PBS-bovine serum albumin (BSA): PBS+1% (w/v) radioimmunoassay grade BSA.
4. PBS-BSA-Tween: PBS + BSA + 0.1% (v/v) Tween-20.
5. High-binding, irradiated polystyrene, 96-well flat-bottomed, microtiter plates
(Immulon 2 HB; Dynex).
6. Polystyrene, 96-well U-bottomed, microtiter plates (Dynex).
7. Photometric microplate reader with 405- and 490-nm filters and software to deter-
mine concentrations of unknowns from standard curves.
8. Secondary antibody: alkaline phosphatase-conjugated goat anti-rabbit IgG
(F[ab']2 fragment of goat antibody).
9. Freshly prepared substrate solution: 0.5 mg/mL disodium p-nitrophenyl phosphate
(104 phosphatase substrate, 5-mg tablet; Sigma) in 10 mL of 1.0 M diethanol-
amine buffer (containing 24.5 mg MgCl2 in 500 mL dH2O, pH to 9.8). Store at
room temperature in amber bottle.
10. ELISA amplification system (Gibco BRL).
11. Peptides synthesized as described in Subheading 3.1.4. with residues added to
(+1, +2, +3), deleted from (–1, –2, –3), and bridging the cleavage site terminus of
the immunizing C1, 2C peptide. For the NC4 and COL2 antibody characteriza-
tions, the NC4 and COL2 synthetic peptides are used.
12. Native, heat-denatured, and collagenase-cleaved types I, II, and III collagen from
species of choice for the COL2-3/4m and C1, 2C antibodies (see Note 1). Puri-
fied native type IX collagen and NC4 domain for the COL2 and NC4 antibodies
(see Note 2).
2.2. Immunoassays
2.2.1. Preparation of the Coating Peptides
for the C1, 2C, COL2, and NC4 Immunoassays
EDTA-Na2, 0.2 g/L NaN3, pH 7.0.
2. Keyhole limpet hemocyanin (KLH): 20 mg (Pierce) reconstituted with 2 mL of

distilled water (dH2O).
3. Coupling reagent: 56.5 mg N-hydroxy-succinimidylbromoacetate/mL of chilled
N,N-dimethylformamide (Sigma).
4. Tris-buffered saline (TBS): 8.78 g/L NaCl, 6.06 g/L Tris-HCl, pH 7.5.
5. Sephadex G-25 column (27 × 2.2 cm) with 103 mL bed volume.
6. C1, 2C (formerly COL2-3/4CShort) synthetic peptide (CGP POH GPQG), COL2
(CGDTGPGVDGRDGY) synthetic peptide containing glycine spacer and N-ter-
minal cysteine and NC4 (CRRPRFPVNSNSNGGNEY) synthetic peptide: used
at 5 mg/100 µL of 0.1 M phosphate buffer.
7. Coating peptides for C1, 2C, and COL2 assays as described in Subheading 3.2.1.
and diluted in TBS, pH 7.5.
8. UV spectrophotometer and quartz cuvets.
2.2.2. Preparation of the Coated Microtiter Plates

1. High-binding, irradiated, polystyrene, 96-well flat-bottomed, microtiter plates
(Immulon 2 HB; Dynex).
For the COL2-3/4m assay, heat-denatured type II collagen (HDC) is prepared
immediately before use. Dissolve native bovine type II collagen (Sigma) in car-
bonate buffer (pH 9.2) to a final concentration of 1 mg/mL in a screw-capped
1.5-mL centrifuge tube. At this concentration it will be partly in suspension. Place
the tube in a water bath at 80ºC for 20 min, but remove it for brief periods of
mixing over the first 2 min, to ensure complete dissolution of the collagen. Dilute
the HDC in carbonate buffer to a final concentration of 40 µg/mL.
2. Carbonate buffer: 1 L of 0.1 M NaHCO3 and 200 mL of 0.1 M Na2CO3. Gradu-
ally add the Na2CO3 to the NaHCO3, with mixing, until the pH is 9.2.
3. TBS: 8.78 g/L NaCl, 6.06 g/L Tris-HCl, pH 7.5.
4. TBS-Tween: TBS + 0.1% (v/v) polyoxyethylenesorbitan monolaurate (Tween-20).
5. TBS-BSA: TBS + 1% (w/v) radioimmunoassay (RIA) grade BSA.
2.2.3. Assaying Samples

2.2.3.1. THE C1, 2C, COL2, AND NC4 ELISAS
1. C2C (for C1, 2C assay), COL2, or NC4 peptides.
2. TBS: 8.78 g/L NaCl, 6.06 g/L Tris-HCl, pH 7.5.
3. TBS-Tween: TBS + 0.1% (v/v) Tween-20.
4. TBS-BSA: TBS + 1% (w/v) RIA grade BSA.
5. TBS-BSA-Tween: TBS + 1% (w/v) BSA + 0.1% (v/v) Tween-20.
6. Polypropylene 96-well U-bottomed microtiter plates (Costar).
7. Affinity-purified C1, 2C, COL2, or NC4 antibodies: as prepared in Subhead-
ing 3.1.3.
8. Secondary antibody: alkaline phosphatase-conjugated goat anti-rabbit IgG
(F[ab']2 fragment of goat antibody; Sigma).
9. ELISA amplification system (Gibco-BRL).

10. Photometric microplate reader with 490-nm filter and software to determine con-
centrations of unknowns from standard curves.
11. Stop solution: 0.3 M H2SO4 (1 mL concentrated H2SO4 slowly added to 59 mL
of dH2O).
2.2.3.2. THE COL2-3/4M ELISA

1. Tris(hydroxymethyl)aminomethane (Tris): prepare a stock solution at 50 mM,
including NaN3, to a final concentration of 0.05% (w/v) and adjust the pH to 7.6
with HCl.
2. PBS: prepare a stock solution of 0.85% (w/v) NaCl, 0.16% (w/v) Na2HPO4,
0.054% (w/v) KH2PO4 and 0.05% (w/v) NaN3. Adjust the pH to 7.4.
3. BSA: prepare a stock solution of 1% (w/v) BSA in PBS and store at 4ºC.
4. PBS/Tween: add Tween-20 (Sigma) to PBS to a final concentration of 0.1%
(v/v).
5. Polypropylene 96-well U-bottomed microtiter plates (Costar).
6. Antibody COL2-3/4m.
7. Peptide CB11B: GKVGPSGA POH GEDGR POH GP POH GPQ.
8. Second antibody: goat anti-mouse IgG labeled with alkaline phosphatase (avail-
able from many sources).
9. Freshly prepared substrate solution: 0.5 mg/mL disodium p-nitrophenyl phos-
phate (104 phosphatase substrate, 5-mg tablet; Sigma in 10 mL of 1.0 M
diethanolamine buffer (containing 24.5 mg MgCl2 in 500 mL dH20, pH to 9.8).
Store at room temperature in amber bottle.
10. Photometric microplate reader with 405-nm filter and software to determine
concentrations of unknowns from standard curves.
2.2.4. Assay Validation

All reagents as listed in Subheading 2.1.4.
2.2.5. Experimental Requirements for the C1,
2C, COL2-3/4m, NC4, and COL2 ELISAs
1. Chymotrypsin digestion solution: 50 mg type VII α-chymotrypsin (Sigma) in
49.25 mL of 50 mM Tris-HCl (6.06 g/L, pH 7.6), plus the general metallo-, cys-
teine, and aspartic protease inhibitors added as 250 µL EDTA (76 mg/mL dH2O),
250 µL iodoacetamide (37 mg/mL dH2O), and 250 µL pepstatin A (2 mg/mL in
95% ethanol), respectively.
2. Proteinase K digestion solution: 50 mg proteinase K (Sigma) plus 250 µL of each
of the protease inhibitors in the chymotrypsin digestion solution, made up to 50 mL
with 50 mM Tris-HCl, pH 7.6.
3. TPCK (irreversible chymotrypsin inhibitor): 8 mg N-tosyl-L-phenylalanine-
chloromethyl ketone (Sigma) in 800 µL of 95% ethanol added dropwise to 19.2 mL
of 50 mM Tris-HCl, pH 7.6.
Fig. 1. Location of cleavage-site neoepitope (C1, 2C) and the hidden denaturation
epitope (COL2-3/4m) in triple-helical type II collagen. Shown above the collagen
molecule is the immunogen (CB11B) that was used to generate the COL2-3/4m mono-
clonal antibody, and below that is the 13-amino acid peptide it specifically recog-
nizes (COL2-3/4m). Below the collagen molecule are the aligned amino acid
sequences of the C termini of the collagenase-generated 3/4 α1-chain fragments of
type II collagen from various species and the collagenase-generated 3/4 α1 and α2
fragments of human type I collagen. A dash indicates an identical amino acid to that
similarly located in the human α1(II) sequence. The peptide sequences used as immu-
nogens for production of the C1, 2C (formerly COL2-3/4Cshort), C2C, and 234CEQ
antineoepitope antibodies are also shown.
3. Methods
3.1. Polyclonal Antibody Production
3.1.1. Preparation of the Immunogens
The amino acid sequence, GPP(OH)GPQG used for the synthesis of the
neoepitope peptide C1, 2C (formerly COL2-3/4Cshort), which corresponds to the
C-terminus of the 3/4 (TCA) fragment of the α1-chains of collagenase-cleaved
triple-helical type II collagen, was based on the published amino acid sequence
for the primary MMP-1 cleavage site in human type II collagen (Fig. 1).
The five C-terminal amino acids in this peptide are identical to the five amino
acids at the C termini of collagenase-cleaved α1 and α2-chain 3/4 fragments of
type I collagen accounting for the crossreactivity of an antibody produced to the
C1, 2C peptide for both collagenase-cleaved type I and II collagens. In an
attempt to produce an antibody specific to type II collagen, a longer peptide,
GAEGPP(OH)GPQG, was synthesized that incorporated the three amino acids
of human type II collagen (in bold) that were N-terminal to the C1, 2C peptide
and designated C2C (formerly COL2-3/4Clong). It is in this region of the col-
lagen molecule that significant differences exist between the α-chains of the
different collagen types, as well as between species for type II collagen (Fig. 1).
This latter feature was utilized in the production of the 234CEQ antibody: the
peptide, GPDGPP(OH)GPQG was synthesized to create an antibody that recog-
nized collagenase-created 3/4 fragments of type II collagen in dogs and horses
(5). The assignment of the third residue in these synthetic peptides as a hydro-
xylated proline, P(OH), is based on the assumption that the prolines in the
Y-position of the repeating Gly-X-Y triplets in collagen α-chains are potential
hydroxylation sites.
For synthesis of the NC4 and COL2 immunogenic peptides, we chose sequences
from within the noncollagenous NC4 domain so that we could monitor the long
form of type IX collagen and the collagenous COL2 domain for total type IX
collagen (4). The cDNA sequences of two distinct forms of the human α1(IX)
collagen chain are known. The type IX collagen COL2 domain sequence
GDTGPGVDGRDG and that for the NC4 domain RRPRFPVNSNSNGGNE were
identified as specific for type IX collagen by reference to the SWISS-PROT pro-
tein sequence database, using MacMolly Tetra software (Soft Gene, Berlin, FRG).
For the production of the COL2-3/4m monoclonal antibody, a peptide was
identified as a hydrophilic domain within the immunogenic cyanogen bromide
peptide 11 (CB11) of the α1(II) chain that satisfied additional criteria as a
suitable immunogen (2). The 21-amino acid sequence GKVGPSGAP(OH)
GEDGRP(OH)GPP(OH)GPQ (CB11B peptide) was synthesized for the immuni-
zation of mice (Fig. 1).
All peptides were synthesized at a 0.25-mmol scale, using standard Fmoc
(9-fluoroenylmethoxycarbonyl) chemistry, on a solid-phase peptide synthesizer.
Crude peptides were purified by reverse phase chromatography. A cysteine was
added to the N termini of the peptides for conjugation to the carrier proteins
OVA, BSA, and KLH. These proteins are commonly used in hapten-protein
conjugates to improve the immunogenicity of the hapten. A tyrosine was added
to the carboxy terminus of the peptide for iodination, if required. The procedure
for conjugation to OVA is as follows:
1. Slowly add (dropwise) 0.2 mL of the coupling reagent to 2.0 mL of chilled OVA,
with stirring, and then warm to 25°C over 30 min.
2. Centrifuge the sample and apply to a previously washed (50 mL of 0.1 M phos-
phate buffer, pH 7.0) Sephadex G-25 column, eluting with phosphate buffer at
1 mL/min and collecting 2-mL fractions.
3. Measure the absorbance at 280 nm in 1-cm quartz cuvets of each fraction, pool
those containing the protein peak, and measure the absorbency of the pooled
sample.
4. Determine the OVA concentration in mg/mL by dividing the absorbency value
by 1.2 (the extinction coefficient of OVA at 1 mg/mL), and adjust OVA to 2–3
mg/mL.
5. Add 4 mg of the activated OVA (about 1.5 mL) to 5 mg of the synthetic peptide
solution (100 µL) and leave at 25°C for 2 h, with occasional mixing, before trans-
ferring to 4°C overnight.
6. Dialyze against 1 L of phosphate buffer (pH 7.0), with three changes over 24 h,
and then against 1 L azide-free PBS with one change over 24 h. Aliquot at 300 µL
and store at –20°C.
7. Confirm effective incorporation by applying coupled OVA-peptide to one lane
and activated OVA to another lane of a 10% sodium dodecyl sulfate-polyacryla-
mide gel electrophoresis (SDS-PAGE) gel and by showing a decrease in electro-
phoretic mobility of the OVA-peptide conjugate on SDS-PAGE relative to the
activated OVA.
3.1.2. Immunization
The particular strain or breed of an animal species may affect the affinity
or specificity of the antibodies produced in that animal. Young (but skeletally
mature), disease-free, female rabbits are the principal source for raising
polyclonal antibodies in the research laboratory. Polyclonals are easily raised
against most antigens and haptens. Young (6–8 wk-old), disease-free, female
Balb/C mice are the principal source of antibody-secreting B cells for producing
monoclonal antibodies in the research laboratory. Monoclonal antibodies are
generally produced after a successfully characterized polyclonal antibody has
been identified, and this chapter will not describe the extensive procedures for
monoclonal antibody production. Thus, the following is the protocol for raising
C1, 2C (formerly COL2-3/4Cshort), NC4, and COL2 polyclonal antibodies:
1. Thaw out one 300-µL aliquot of peptide-OVA conjugate, add 600 µL of azide-
free PBS and mix (~1 mg conjugate/mL).
2. Take up the peptide-OVA conjugate into a glass syringe and emulsify in 900 µL
of complete Freund’s adjuvant (CFA) by forcing through a 20-gage needle sev-
eral times until a drop of emulsion retains its shape when added to the surface of
a beaker of water (final concentration ~0.5 mg/mL).
3. Inject two rabbits intramuscularly (in both hind limbs) with 0.5 mL of emulsion
(~ 200 mg of conjugate) divided between two different sites.
4. Give booster injections of similar quantities of peptide-OVA emulsified with
incomplete Freund’s adjuvant (IFA) intramuscularly every 3–4 wk.
5. Perform test bleeds and determine antibody titers by immunoassay (see Sub-
heading 3.1.4.) after the second booster. If the titers are good, exsanguinate the
rabbits by cardiac puncture and collect serum for antibody purification.
3.1.3. Antibody Purification

A wide variety of methods can be used to purify antibodies. The choice of
purification will depend on a number of factors, including the species in which
the antibody was produced, the type of antibody (monoclonal or polyclonal),
the class and subclass of the antibody, the source (serum, ascitic, or tissue cul-
ture fluids), and the intended use of the antibody.
Polyclonal antibodies are usually of the IgG class, and generally only 10% of
that IgG is specific for the immunizing antigen. A two-step approach is often
used to purify rabbit antibodies consisting of ammonium sulfate precipitation
followed by affinity chromatography. A 50% saturated solution of ammonium
sulfate is used, as most serum components will not precipitate in this range. The
resulting crude IgG preparation is then digested with pepsin to yield antibody
fragments with the two antigen-binding domains still bound together, called
F(ab')2 fragments. This process removes the Fc portions of the IgG molecules
that can produce nonspecific binding in immunochemical assays. Finally the
F(ab')2 fragments specific for the antigen of interest are purified by affinity chro-
matography on a column containing the immunizing peptide coupled through
its added cysteine terminal end to a sulfur-linking gel.
The affinity purification procedure used for the F(ab')2 fragments of the C1,
2C (formerly COL2-3/4Cshort), COL2, and NC4 polyclonal IgG antibodies is
as follows:
1. Centrifuge the volume of serum from which the antibody is to be purified for 30
min at 3000g and place the supernatant in a beaker with a stirring bar.
2. With the beaker on a magnetic stirrer, slowly add to the serum an equal volume
of a saturated solution of ammonium sulfate and incubate at 4°C overnight.
3. After centrifuging for 30 min at 3000g, resuspend the pellet in 0.5 vol (of initial
volume) of PBS and dialyze with three changes of PBS overnight.
4. Remove an aliquot of the antibody solution to determine the IgG concentration
(mg/mL) by dividing the absorbance at 280 nm by 1.35 (at 1 mg/mL) and then
adjust the concentration to 10–30 mg/mL, by adding PBS.
5. To the crude IgG preparation, first add one-third vol of 0.8 M sodium acetate
buffer (pH 4.0), followed by 2% (w/w) pepsin, dissolved in a drop of distilled
water, and incubate overnight at 37°C.
6. Adjust the pH of the solution to 8.6 with NaOH to inactivate the pepsin and then
back to neutrality with acetic acid.
7. Add an equal volume of a saturated solution of ammonium sulfate, centrifuge
the suspension, and discard the supernatant. Dissolve the precipitate in 3 mL of
0.14 M phosphate buffer (pH 8.0) and desalt with Econo-Pac 10DG columns
containing Bio-Gel P-6 gel, according to the manufacturer’s instructions

(Gibco-BRL).
8. Separate the F(ab')2 fragments from the remaining intact antibodies and Fc frag-
ments by passing the solution through a protein A column (10–20 mg of IgG
per mL wet beads). Wash the beads with 10 column vol of 100 mM Tris-HCl
(pH 8.0), 10 column volumes of 10 mM Tris (pH 8.0) and elute the F(ab')2
fragments with 0.1 M glycine-HCl (pH 3.0). Collect 1-mL aliquots in tubes
containing 50 µL of 1 M Tris (pH 8.0) and gently mix to avoid frothing. Iden-
tify the IgG-containing fractions by absorbance at 280 nm, pool, and reread the
absorbance at 280 nm.
9. Prepare an affinity column with the C1, 2C, NC4, or COL2 peptides coupled to
1 vol of SulfoLink coupling gel (1 mL peptide solution/mL gel) according to the
manufacturer’s instructions (Pierce) and wash with 3 vol of PBS.
10. Apply 1 vol of the F(ab')2 solution to the column and allow to enter the gel com-
pletely before closing the bottom of the column, then add 0.4 vol of buffer, and
leave the column at room temperature for 1 h.
11. Wash the column with 8 vol of PBS and elute the C1, 2C, NC4, or COL2 F(ab')2
fragments with 4 vol of 0.1 M glycine-HCl (pH 3.0). Collect 1-mL fractions in
tubes containing 50 µL of 1 M Tris-HCl (pH 8.0), and read the absorbance for
each at 280 nm. Pool the fractions containing the F(ab')2 fragments, dialyze over-
night in PBS, reread the absorbance, and then store in 0.5-mL aliquots at –20°C.
3.1.4. Antibody Characterization

A true anti-neoepitope antibody to MMP-cleaved collagens should recog-
nize the exact sequence of amino acids representing the newly created N or C
termini of the collagen α-chain fragments. To determine this specificity, pep-
tides are synthesized with up to three amino acid residues either added to (+1,
+2, +3) or removed from (–1, –2, –3) the end of the immunizing peptide corre-
sponding to the cleaved terminus: a peptide can also be prepared that repre-
sents the sequence bridging the cleavage site of the native collagen α-chain.
The antibody must recognize the MMP-cleaved collagen type(s) for which it
was designed. An antibody for denatured collagen must recognize sequences
in “unwound” collagen α-chain fragments that are not accessible for antibody
binding in the intact (native) triple helical molecule. It must also recognize the
denatured collagen type for which it was designed. The antibodies to type IX
collagen should recognize the exact sequence of amino acids representing the
NC4 or COL2 domains from which the immunogens were derived.
3.1.4.1. PREPARING THE PLATES FOR IMMUNOASSAY

1. For the coating of the C1, 2C immunoassay plates, the peptide, synthesized to
include two glycine spacers and an added N-terminal cysteine (Fig. 1), is conju-
gated to BSA using the methodology described in Subheading 3.1.1., with the
exception that the absorbance of the pooled fractions with the activated BSA is
divided by 0.7 (the extinction coefficient at 1 mg/mL) to determine the BSA
concentration.
2. For the coating of the NC4 and COL2 immunoassay plates, peptides are used that
include a N-terminal cysteine.
3. For the coating of the COL2-3/4m immunoassay plates, HDC is used.
4. Dilute the assay peptide (C2C-BSA conjugate, NC4 or COL2) to 20 µg/mL or
the HDC (COL2-3/4m assay) to 40 µg/mL in 0.1 M carbonate buffer (pH 9.2), to
promote noncovalent adsorption of the conjugate, and add 50 µL to each well of
Immulon-2, 96-well flat-bottomed microtiter plates.
5. After overnight incubation at 4°C, wash the plates three times with PBS-Tween,
block the noncoated binding sites with 150 µL/well of PBS-BSA for 30 min at
room temperature and then wash once with PBS-Tween. Plates can then be stored,
covered with plastic wrap, for up to 6 mo at 4°C.
3.1.4.2. ANTIBODY TITER

1. Add serial dilutions (1:10–1:1280) of the antibody preparations as 50-µL
aliquots, in duplicate, to wells of the coated plates; after a 90-min incubation at
37°C, wash the plates with PBS-Tween. Seal the plates in paraffin sheets for all
incubations.
2. Dilute the secondary antibody conjugate according to the manufacturer’s recom-
mendations in PBS-BSA-Tween and add at 50 µL/well for 1 h at 37°C.
3. Wash the plates three times with PBS-Tween and once with dH20, and then add
the alkaline phosphatase substrate at 50 µL/well for 20–30 min at 37°C. Read the
absorbance at 405 nm with a microplate photometer.
3.1.4.3. ANTIBODY SPECIFICITY

1. Precoat 96-well U-bottomed microtiter plates with PBS-BSA at 100 µL/well for
30 min at room temperature and wash once with PBS-Tween.
2. To act as nonspecific binding (NSB) wells, add 50 µL of PBS-BSA and 50 µL
of PBS-BSA-Tween to each of four wells. To all other wells add 50 µL of the
appropriate dilution of antibody preparation (see Note 3).
3. As the immunoassays are inhibition ELISAs, add 50 µL of PBS-BSA to each of
another four wells containing 50 µL antiserum, to determine maximum binding
(MB) of the antibody in the absence of the inhibitory peptides.
4. To each of remaining test wells add, in duplicate, 50 µL of serial dilutions of
either the C1, 2C, or C2C peptides, the addition (+1, +2, +3) and deletion (–1,
–2, –3) peptides described above, the peptide bridging the collagenase cleavage
site in native type II collagen α-chains, or native (triple-helical), heat-dena-
tured (80°C for 20 min) and collagenase-cleaved type I, II, and III collagens
(see Note 1). For the NC4 and COL2 immunoassays, add, in duplicate, 50-µL
dilutions of NC4 or COL2 peptides, NC4 domain, or type IX collagen (see
Note 2). For the COL2-3/4m ELISA, add, in duplicate, 50-µL dilutions of
CB11B peptide or HDC.
5. After 1 h at 37°C, transfer 50 µL from each well to the equivalent wells of plates
precoated, as described above, with the one exception that the C2C-BSA conju-
gate is diluted at 50 ng/well in PBS (pH 7.2) for the C1, 2C assay.
6. Incubate the plates for 30 min at 4°C, wash three times with PBS-Tween, and
then add the secondary antibody conjugate, at the dilution recommended by the
manufacturer in PBS-BSA-Tween, for 1 h at 37°C.
7. After a further three washes with PBS-Tween and dH20, use an ELISA ampli-
fication system kit, according to manufacturer’s instructions (Gibco-BRL) or
50 µL/well of alkaline phosphatase (AP) substrate solution (Sigma).
8. Read the absorbance for each well at the appropriate wavelength (490 nm for the
amplification kit and 405 nm for the AP substrate) and subtract the mean of the
NSB wells from the values for each of the other wells to obtain a corrected absor-
bance. Calculate the percentage binding for each peptide from their corrected
absorbance values relative to the mean absorbance of the four MB wells, which
represent 0% inhibition.
corrected absorbance (absorbance – mean absorbance of nonspecific well) × 100

% binding =
mean absorbance of maximum binding wells
Calculate the percentage inhibition by subtracting the percentage binding from 100.
3.2. Immunoassays
3.2.1. Preparation of the Coating Peptides
for the C1, 2C, and COL2 Immunoassays
To improve the sensitivity of the C1, 2C immunoassay, which is of particu-
lar importance when assaying biological fluids, the C1, 2C peptide is conju-
gated to KLH. A similar approach was used for the COL2 immunoassay. This
is not necessary for the NC4 assay. The KLH protein is commonly used in
hapten-protein conjugates to enhance the noncovalent binding of a hapten to a
solid support, such as the wells of microtiter plates used in immunoassays, and
to ensure that the peptide is “exposed” for antibody binding. A cysteine was
added to the N-termini of both the C1, 2C and COL2 peptides for conjugation
to the KLH carrier protein.
The procedure for conjugation of the C1, 2C and COL2 peptides to KLH for
use as the coating peptide in the immunoassays is as follows:
1. Reconstitute 20 mg of KLH with 2 mL dH20 and dialyze overnight against 0.1 M
phosphate buffer (pH 7.0).
2. Slowly add (dropwise) 0.2 mL of the coupling reagent to 2.0 mL of chilled KLH,
with stirring, warm to 25°C over 30 min, and then cool on ice.
3. Repeat step 2, but do not cool after warming.
4. Centrifuge the sample and apply the supernatant to a previously washed (50 mL
of 0.1 M phosphate buffer, pH 7.0) Sephadex G-25 column, eluting with phos-
phate buffer at 1 mL/min and collecting 2-mL fractions.
5. Measure the absorbency at 280 nm in quartz cuvets of each fraction, pool those
containing the first protein peak (bluish gray color), and measure the absorbency
of the pooled sample.
6. Determine the [KLH] in mg/mL by dividing the absorbency value by 1.5 (the
extinction coefficient of KLH at 1 mg/mL) and adjust [KLH] to 2–3 mg/mL.
7. Add 4 mg of the activated KLH (about 1.5 mL) to 5 mg of the synthetic peptide in
solution (100 µL) and leave at 25°C for 2 h, with occasional mixing, before trans-
ferring to 4°C overnight.
8. Dialyze against 1 L of phosphate buffer (pH 7.0), with three changes over 24 h,
and then against 1 L TBS with one change over 24 h. Aliquot at 300 µL and store
at –20°C.
9. Confirm effective incorporation by applying coupled KLH-peptide to one lane
and activated KLH to another lane of a 10% SDS-PAGE gel and by showing a
decrease in electrophoretic mobility of the KLH-peptide conjugate on SDS-
PAGE relative to the activated KLH.
3.2.2. Preparation of the Coated Microtiter Plates

The immunoassays are inhibition ELISAs, wherein standards and samples
are preincubated with the primary antibody, and the more of the epitope that is
present in each, the less antibody that is left to bind to the solid phase epitope,
i.e., the heat-denatured type II collagen (HDC) or the C1, 2C-KLH, COL2-
KLH, and NC4 peptides that are coating the immunoassay plates. Here is the
procedure for coating the microtiter plates to be used in these inhibition
ELISAs:
1. Add to all but the 36 perimeter wells of 96-well flat-bottomed Immulon 2
microtiter plates, 50 µL/well (5 ng/well) of the C1, 2C-KLH conjugate or NC4
peptide diluted to 100 ng/mL in TBS (pH 7.5) and incubate overnight at 4°C.
For the COL2 assay, add 50 µL/well (2 ng/well) of the COL2 peptide coupled
to KLH and diluted to 40 ng/mL in TBS (pH 7.5). For the COL2-3/4m assay,
add 50 µL/well (2 µg/well) of the HDC diluted to 40 µg/mL in carbonate buffer
(pH 9.2). The exclusion of the outer wells is because of the potential for exces-
sive evaporation from these wells during incubation (Fig. 2).
2. After washing three times with TBS-Tween, block the plates with 100 µL/well of
TBS-BSA for 30 min at 25°C and then wash three times with TBS-Tween.
3.2.3. Assaying Samples

The following is the procedure for all the ELISAs, with differences between
the assays noted:
1. Prepare standards for the C1, 2C ELISA by diluting a 1 mg/mL stock solution of
C2C peptide to 10, 2.5, 0.63, 0.16, 0.04, 0.01 and 0.0025 µg/mL in TBS-BSA.
2. Prepare standards for the NC4 ELISA by diluting a 1 mg/mL stock solution of
NC4 peptide to 100, 50, 25, 12.5, 3.12, 1.56, 0.78, and 0.39 ng/mL in TBS-BSA.
Fig. 2. Microtiter plate template for the C1, 2C ELISA. The perimeter wells cor-
responding to columns 1 and 12 and to rows A and H are not used. All blanks,
standards, and samples are assayed in duplicate. Symbols shown represent nonspe-
cific binding (NSB) wells with no C1, 2C antibody added, maximum binding (MB)
wells with no inhibiting C2C peptide added, standard wells (S1–S7) with 10, 2.5,
0.63, 0.16, 0.04, 0.01, and 0.0025 µg of inhibiting C1, 2C peptide/mL of TBS-BSA,
and sample wells (1–21). Similar templates are employed for the NC4, COL2, and
COL2-3/4m immunoassays, with different concentrations of standards used, as indi-
cated in Subheading 3.2.3.
3. Prepare standards for the COL2 ELISA by diluting a 1 mg/mL stock solution of
COL2 peptide to 10000, 1000, 100, 10, 1, 0.1, and 0.01 ng/mL in TBS-BSA.
4. Prepare standards for the COL2-3/4m ELISA by diluting a 1 mg/mL stock solu-
tion of COL2 peptide to 8, 6, 4, 3, 2, 1, 0.5, and 0.25 µg/mL in TBS-BSA.
5. Add each standard at 50 µL/well, in duplicate, to 96-well U-bottomed polypro-
pylene plates, as outlined in the template shown as Fig. 2, again ignoring all
outside wells. The duplicate NSB wells consist of 50 µL TBS-BSA and 50 µL
TBS-BSA-Tween, and the duplicate maximum binding (MB) wells contain
50 µL TBS-BSA and 50 µL of the antibody preparation, at a dilution in TBS
(pH 7.5), as previously determined (see Note 3).
6. Add samples at 50 µL/well, in duplicate, to the remaining wells, and then add to
each of the wells 50 µL of the diluted antibody preparations, before covering the
plates with paraffin sheets and incubating at 37°C for 1 h.
7. Using a multichannel pipetor, transfer 50 µL from each well of the polypropy-

lene plates to the respective well of the HDC, C1, 2C-KLH, NC4, or COL2-KLH
coated plates. Incubate these plates for 30 min at 4°C and then wash three times
with TBS-Tween.
8. Dilute the secondary antibody conjugated to AP in TBS-BSA-Tween, as recom-
mended by the manufacturer or optimized in the specific assay. Add 50 µL/well
and incubate for 1 h at 37°C. Wash the plates three times with TBS-Tween and
three times with dH2O.
9. Use a two-step ELISA amplification system according to the manufacturer’s in-
structions (Gibco-BRL) or 50 µL/well of AP substrate (COL2-3/4m assay), and
after the color development is stopped with 0.3 M H2SO4, read the absorbance at
either 490 nm (amplification system) or 405 nm (AP substrate).
10. Determine the concentration of the C1, 2C, NC4, COL2, and COL2-3/4m epitopes
in each sample from a log/linear plot of concentration (µg/mL) vs corrected
absorbance (sample absorbance – mean of nonspecific binding absorbance) for
the standards. Use nonlinear regression to establish a sigmoidal dose–response
curve with variable slope for determination of the concentrations of the unknowns.
Typical standard curves for these immunoassays are shown in Fig. 3.
3.2.4. Assay Validation

Assays that are developed for use in clinical sciences must be validated.
There are guidelines published by the International Federation of Clinical
Chemistry.
The features of any assay that need to be validated are specificity, sensitiv-
ity, precision (within run and between run), analytic recovery, and identity of
calibrant and unknown sample.
3.2.4.1. ANTIBODY SPECIFICITY

Antibody specificity refers to the ability of an antibody to produce a mea-
surable response for the analyte of interest, whereas crossreactivity measures
the response of the antibody to other substances. The polyclonal C1, 2C anti-
body has selective high affinity for both the immunizing C1, 2C peptide and
collagenase (MMP-1 and MMP-13)-cleaved human type II collagen 3/4
α-chain fragments, when compared on a molar basis (1). The antibody demon-
strates a lower affinity for similarly cleaved human type I collagen α-chains.
There is negligible binding to uncleaved triple-helical and heat-denatured
human types I and II collagen, as well as intact or cleaved type III collagen
α-chains. The materials required for these procedures are listed in Subheading
2.1.4. and Note 1.
The NC4 and COL2 antibodies have selective high affinity for both the
immunizing peptide and type IX collagen extracted from tissue (3). There was
no cross reactivity with collagens types I, II, III, V, VI, and X. Antisera speci-
Fig. 3. Standard curves for NC4 and COL2 domain assays for type IX collagen
α1(IX) chain are shown together with those for type II collagen α1(II) chain intact and
cleaved, namely, COL2-3/4m and C1, 2C, respectively.
ficities were also characterized with inhibition ELISA using purified NC4
domain (for NC4 antiserum), pepsin digested type IX collagen (for COL2
antiserum), and collagen types I, II, III, V, VI, and X as competing antigens.
Only the corresponding peptides, pepsin-digested type IX collagen (for COL2
antiserum) and purified NC4 domain (for NC4 antiserum), were inhibitory,
whereas collagen types I, II, III, V, VI, and X had no effect (4). There was no
crossreactivity with collagen types I, II, III, V, VI, and X.
The COL2-3/4m antibody reacts with denatured type II collagen but not
denatured types I, III, or X collagen. It crossreacts with the α-3 chain of type
XI collagen, as the sequence is identical to that of the α1(II)-chain. It does not
react with native (triple-helical) type II collagen.
3.2.4.2. SENSITIVITY
The sensitivity of an assay is the lowest dose at which there is a significant
difference in response from zero. This is generally determined by the upper 2
or 3 SD limit in a zero standard study. The minimum detection limit of the C1,
2C assay is 4.9 nM.
3.2.4.3. PRECISION
Within-run (intra-assay) and between-run (interassay) precision is deter-
mined by assaying different samples that cover the range of detectable levels
of epitope, in a series of runs. For within-run precision, a single sample is
assayed in multiple replicates on the same plate (n = +20). The coefficient of
variation (% CV) is determined by dividing the SD by the mean of the concen-
tration values obtained and then multiplying by 100. The between-run preci-
sion is determined from the same samples assayed on multiple plates (n = +5),
and the % CV is determined from the SD of the plate means and the overall
concentration mean. Values obtained for collagenase-cleaved human type II
collagen and culture medium from IL-1–stimulated bovine articular cartilage
explants range from CVs of 4.6–8.3% for within-run precision to CVs of 5.4–
6.9% for between-run precision for the C1, 2C immunoassay.
3.2.4.4. ANALYTIC RECOVERY
“Spike recovery” is determined by the addition of known quantities of the
purified peptide into fluid samples with different levels of endogenous epitope.
Typical results for synovial fluid, urine, and cartilage explant culture medium
range from 89 to 110 % recovery for the C1, 2C immunoassay.
Linearity is shown by diluting samples serially and comparing the observed
values with those expected. Typical recovery rates of 94–111% have been noted
for cleaved type II collagen and culture medium from IL-1–stimulated articu-
lar cartilage explants using the C1, 2C immunoassay.
3.2.4.5. IDENTITY OF ANALYTE AND CALIBRANT
The demonstration of parallelism between the dose response curves of the
calibrant and the unknown sample proves identity (4). The C1, 2C, NC4, COL2,
or CB11B peptides and the sample (cleaved type II collagen for C1, 2C, type IX
collagen for COL2, heat-denatured type II collagen for COL2-3/4m, NC4
domain, culture medium, serum, synovial fluid, urine, and so on) are each pro-
gressively diluted and assayed separately as described above in Subheading
3.2.3. The results for each are plotted on the same graph of concentration (dilu-
tion) vs % inhibition, and the curves for each are compared for parallelism (1–3).
3.2.5. Experimental Studies Using the C1,
2C, NC4, COL2, and COL2-3/4m ELISAs
The C1, 2C ELISA can be used to assess the degree of MMP activity, spe-
cifically for the collagenases MMP-1, MMP-8, and MMP-13, in terms of
cleaved product, in any system in which there is type I and/or II collagen. It can
be used to evaluate the effect of MMP inhibitors on the generation of collage-
nase-cleaved collagen products in vivo and in vitro (1,5–7). To evaluate the
levels of collagenase-cleaved collagen in situ, tissue is digested with α-chy-
motrypsin, which releases cleaved and denatured collagens from the extracel-
lular matrix and preserves the recognition of the neoepitope by the C1, 2C
antibody (1,7,8).
The NC4 assay can be used to determine the content of the long form of type
IX collagen, and the COL2 assay can be used to determine total type IX collagen
content in any system in which type IX collagen is present (4). To evaluate the
levels of these collagens in situ, the tissue is digested with α-chymotrypsin and
proteinase K, which releases cleaved and denatured collagens from the extracel-
lular matrix and preserves the recognition of the epitopes.
The COL2-3/4m assay can be used to determine the absolute amount of type
II collagen and the proportion of denatured type II collagen in arthritic articu-
lar cartilage as well as intervertebral disks (2,9–11).
The protocol for digestion of articular cartilage and extraction of degraded
collagen is as follows:
1. Mince articular cartilage and incubate 10–50 mg overnight at 37°C with 500 µL
of 1 mg/mL α-chymotrypsin digestion solution (containing protease inhibitors,
as described above) in 1.5-mL centrifuge tubes with a V-shaped bottom and a
screw-cap lined with an O-ring.
2. Incubate the digest for 20 min at 37°C with 200 µL (160 µg/mL final concentra-
tion) of TPCK.
3. For the NC4 and COL2 ELISAs, digest with 500 µL of proteinase K digestion
solution containing protease inhibitors. Centrifuge the samples and remove the
supernatants for assaying with the respective ELISA, as described in Subhead-
ing 3.2.3.
4. For the COL2-3/4m assay, using an appropriate pipete, transfer all the α-chy-
motrypsin cartilage-extract solution (CE) to a new screw-capped 1.5-mL centri-
fuge tube, taking care to leave all the undigested cartilage residue (CR) in the
original tube. Mix the CE and then transfer 300 µL to another screw-capped 1.5-
mL centrifuge tube for further digestion of the extract (DE). Store the remaining
CE at 4ºC.
5. To the CR add 500 µL of 1 mg/mL proteinase K (including proteinase inhibitors,
as described in step 3). Put the cap on tightly and tap the tube until all the carti-
lage is submerged in the solution. To the DE add 100 µL of 1 mg/mL proteinase
K and put the cap on.
6. Incubate CR and DE overnight in an oven or water bath, at 56ºC. Then increase
the temperature to 100ºC and allow the samples to boil for 10–15 min, to inacti-
vate any remaining proteinase K. See Note 5 for a discussion of problems
encountered with digesting cartilage residues.
7. Assay CE, DE, and CR for epitope CB11B, by inhibition ELISA, as described in
Subheading 3.2.3. Appropriate dilutions of CE, DE, and CR (see Note 6) should
be prepared in Tris-HCl and 50 µL added to duplicate wells already containing
antibody in the COL2-3/4m ELISA.
4. Notes
1. If the collagens for the species of interest cannot be purchased, they can be puri-
fied from tissues of that species. Type II collagen can be purified from articular
cartilage of the species of choice, by pepsin digestion and differential salt pre-
cipitation (12), and types I and III collagen can be prepared by pepsin digestion
and differential denaturation/renaturation (13).
a. For cleavage by the collagenases (MMP-1, MMP-8, and/or MMP-13), these
lyophilized collagens are dissolved separately in 0.5 M acetic acid and diluted
to 2.5 mg/mL in a digestion buffer consisting of 50 mM Tris-HCl, 10 mM
CaCl2, 0.5 M NaCl, 0.01% Brij 35, and 0.02% NaN3, pH 7.6.
b. The proenzymes are activated with 2 mM (final concentration) p-aminophe-
nylmercuric acetate (APMA) in the same digestion buffer for 90 min at 37°C.
c. The activated enzyme solution is added to each of the collagen solutions at
molar ratios of from 1:5 to 1:10, the pH is adjusted to 7.5 with NaOH (if
necessary), and the samples are incubated at 30°C for 24 h.
d. The MMPs are inactivated with 20 mM (final concentration) EDTA, and
cleavage is confirmed by SDS-PAGE under denaturing conditions using 10%,
1-mm-thick, mini-Protean gels (Bio-Rad).
e. To confirm immunoreactivity of the cleavage-site neoepitope antibody, the
electrophoretically separated collagenase-cleaved α-chain fragments are
transferred to a nitrocellulose membrane.
f. After blocking in PBS-3% BSA, the membrane is incubated overnight with
the anti-neoepitope antibody.
g. After washing with PBS-BSA-Tween, secondary antibody conjugate diluted
in PBS-3% BSA-Tween is added for 1 h at 25°C, the membranes are washed
three times in PBS-BSA-Tween and three times in dH20, and substrate solu-
tion is added (5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazo-
lium) from a kit (Bio-Rad).
h. After the color has developed for the immunostained bands, and before
excess background staining (10–20 min), the substrate is removed, and the
membranes are washed with dH20 before drying on filter paper.
2. Isolation of native type IX collagen and the NC4 domain.
a. Slices (40 g wet weight) of fetal bovine epiphyseal cartilage can be extracted
for 48 h at 4°C with constant stirring in 4 M guanidine HCl in 100 mM sodium
acetate, pH 6.0, to remove proteoglycans.
b. The supernatant can be removed by centrifugation at 2000g for 15 min.
c. Residual cartilage is washed with 0.5 M acetic acid and then extracted with
1 M NaCl, 0.05 M Tris-HCl, pH 7.4, at 4°C for 48 h (4).
d. The guanidine HCl and NaCl extractions also contain protease inhibitors
2 mM phenylmethanesulfonyl fluoride (PMSF), 2 mM EDTA-Na2, 5 mM
benzamidine, and 10 mM N-ethylmaleimide.
e. Intact type IX collagen lacking the chondroitin sulfate chain is recovered from
the 1 M NaCl salt neutral extract by precipitation with ammonium sulfate at
30% saturation.
f. The precipitate is redissolved in 0.5 M acetic acid, and native type IX col-
lagen can be enriched in the supernatant after precipitating other molecules
with 0.7 M NaCl.
3. The NC4 domain is purified separately (4).
a. Briefly, proteoglycans are removed from 40g of bovine epiphyseal cartilage
slices as described in step 2.
b. After centrifugation at 2000g for 15 min, the residue is resuspended at 4°C in
30 mL of 50 mM Tris-HCl, pH 7.2, 5 mM CaCl2, with inhibitors (as specified
above).
c. After centrifuging and washing the residue by resuspending it in the same
buffer several times, 4000 U of collagenase (Clostridiopeptidase A, Sigma)
are added, and the sample is incubated at room temperature for 6 h.
d. After centrifugation at 4000g for 60 min at 4°C, the supernatant is chroma-
tographed on a 3 × 100-cm column of 8% agarose (Bio-Rad A 1.5 m).
e. The column is equilibrated with 50 mM Tris-HCl, pH 7.6, and 2 M urea.
f. Two-milliliter fractions are collected and analyzed by PAGE and Western
blotting with anti-NC4 antiserum.
g. Fractions containing the NC4 domain are pooled, dialyzed against 5 mM Tris-
HCl, pH 7.6, and 0.2 M urea, and then lyophilized.
4. Checkerboard analysis can be used for titration of both the antigen and antibody
to determine the optimum concentration of antigen to coat the microtiter plate
wells and the optimum dilution of primary antibody to allow maximum sensitiv-
ity and minimal background in the ELISA.
a. The peptide-carrier conjugate is diluted across the plate from left to right,
with the last column of the wells containing no antigen, and allowed to adsorb
to the plate.
b. The primary antibody is then diluted across the plate from top to bottom,
thereby obtaining a checkerboard titration of antigen against antibody.
c. After incubation at room temperature for 1 h, enzyme-conjugated secondary
antibody is added, and incubated at 37°C for 1 h, followed by the addition of
the enzyme substrate, and the absorbances are read.
d. A graph is made showing a plot for each antibody dilution of antigen dilu-
tion vs optical density (OD), and from this the background of the plate and
the nonspecific adsorption of conjugate, as well as the plateau region of
maximum antigen binding to the plate (plate saturation), can be determined.
The anti-collagen antibody (COL2-3/4m) must also be used at an appro-
priate dilution. If it is too dilute, then the MB value will be too low and inhi-
bition will be difficult to detect. If it is too concentrated, then higher concen-

trations of epitope will be needed to inhibit its binding to HDC on the ELISA
plate. For optimal conditions, the antibody should be bound to HDC at a range
of dilutions over a 30-min incubation time at ambient temperature. After
washing the plate, bound antibody can be detected using the second antibody/
AP detection system. A plot of log-dilution against OD405 will approximate to
a straight line. The dilution producing an OD405 that is about 40% of maximal
OD is optimal for use in the assay. However, note that when the antibody is
added to the preculture plate, it is diluted 1:1 with sample (or buffer), so it
should be prepared initially at 2X the chosen concentration.
5. Total CB11B in the CE compartment represents the amount of denatured type
II collagen. The digestion of CE with proteinase K to generate DE is a means of
degrading any native-type II collagen extracted intact by α-chymotrypsin, thus
making it available for assay. The amount of CB11B in DE should be equal to
or greater than CB11B in CE. Therefore total CB11B epitope in the cartilage =
CB11B in DE + CB11B in CR. Calculate type II collagen degradation as:
% denaturation = (CB11B in CE/total CB11B) × 100
6. Digestion of cartilage residues by proteinase K generally results in complete solu-
bilization of the tissue, which is necessary for an accurate estimate of the total
collagen content. However, in some instances proteinase K fails to solubilize the
tissue. We have found two different reasons for this. The first is if too much
cartilage was added to the centrifuge tube. In this case the remaining undigested
residue can be transferred to fresh proteinase K. The second is that calcified car-
tilage is resistant to proteolysis. This problem is most common with cartilage
from animals such as rabbits or guinea pigs because the uncalcified tissue layer is
very thin. The way to resolve this problem is to increase the concentration of
EDTA in the α-chymotrypsin and proteinase K solutions to 400 mM. (The tetra-
sodium salt of EDTA must be used for this.)
7. The appropriate sample dilutions will vary enormously depending on the type of
cartilage, amount of tissue used, and extent of collagen degradation. However, as
a general guideline, the CE and DE samples may require no dilution, or they may
need to be diluted to about 1:4. The CR samples will always require diluting,
often by as much as 1:40 or 1:80. Clearly these values will be even more varied
when other tissues are being extracted and assayed for their specific collagens.
Acknowledgments
Financial Support was provided by the Shriners Hospitals for Children,
Canadian Institutes of Health Research, National Institutes of Health (to
ARP), and the Fonds de la Recherche en Santé du Quebec (to RCB).
References
1.
1 Billinghurst, R. C., Dahlberg, L., Ionescu, M. et al. (1997) Enhanced cleavage of
type II collagen by collagenases in osteoarthritic articular cartilage. J. Clin. Invest.
99, 1534–1545.
2.
2 Hollander, A. P., Heathfield, T. F., Webber, C., et al. (1994) Increased damage to
type II collagen in osteoarthritic articular cartilage detected by a new immunoas-
say. J. Clin. Invest. 93, 1722–1732.
3.
3 Nelson F, Dahlberg L, Laverty, S., Reiner et al. (1998) Evidence for altered
synthesis of type II collagen in patients with osteoarthritis. J. Clin. Invest. 102,
2115–2125.
4.
4 Mwale, F., Billinghurst, C., Wu, W., et al. (2000) Selective assembly and remod-
elling of collagens II and IX associated with expression of the chondrocyte hyper-
trophic phenotype. Dev. Dyn. 218, 648–662.
5. Billinghurst, R. C., Buxton, E. M., Edwards, M. G., McGraw, M. S., and
McIlwraith, C. W. (2001) Use of an antineoepitope antibody for identification of
type-II collagen degradation in equine articular cartilage. Am. J. Vet. Res. 62,
1031–1039.
6.
6 Dahlberg, L., Billinghurst, R. C., Manner, P., et al. (2000) Selective enhance-
ment of collagenase-mediated cleavage of resident type II collagen in cultured
osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collage-
nase 1 (matrix metalloproteinase 1). Arthritis Rheum. 43, 673–682.
7.
7 Billinghurst, R. C., O’Brien, K., Poole, A. R., and McIlwraith, C. W. (1999) Inhi-
bition of articular cartilage degradation in culture by a novel nonpeptidic matrix
metalloproteinase inhibitor. N. Y. Acad. Sci. 878, 594–597.
8. Billinghurst, R. C., Wu, W., Ionescu, M., et al. (2000) Comparison of the degra-
dation of type II collagen and proteoglycan in nasal and articular cartilages by
interleukin-1 and the selective inhibition of type II collagen cleavage by collage-
nase. Arthritis Rheum. 43, 664–672.
9.
9 Laverty, S., O’Kouneff, S., Ionescu, M., et al. (2002) Excessive degradation of
type II collagen in articular cartilage in equine osteochondrosis. J. Orthop. Res.
20, 1282–1289.
10.
10 Hollander, A. P., Heathfield, T. F., Liu, J. J., et al. (1996) Enhanced denaturation
of the α1(II) chains of type-II collagen in normal adult human intervertebral discs
compared with femoral articular cartilage. J. Orthop. Res. 14, 61–66.
11.
11 Antoniou, J., Steffen, T., Nelson, F., et al. (1996) The human intervertebral disc.
Evidence for changes in the biosynthesis and denaturation of the extracellular
matrix with growth, maturation, ageing and degeneration. J. Clin. Invest. 98,
996–1003.
12.
12 Squires, G., Okouneff, S., Ionescu, M., and Poole, A. R. (2003) Pathobiology of
focal lesion development in aging human articular cartilage reveals molecular
matrix changes characteristic of osteoarthritis. Arthritis Rheum. 48, 1261–1270.
13.
13 Miller, E. J. (1971) Isolation and characterization of a collagen from chick carti-
lage containing three identical achains. Biochemistry 10, 1652–1659.
14. ChandraRajan, J. (1978) Separation of type III collagen from type I collagen and
pepsin by differential denaturation and renaturation. Biochem. Biophys. Res.
Commun. 83, 180–186.
Apoptosis in Cartilage and Chondrocytes 275
19
Detection of Apoptosis in Cartilage
In Situ and in Isolated Chondrocytes
Darryl D. D’Lima, Klaus Kuhn, and Martin K. Lotz
Summary
Apoptosis is a form of programmed cell death conceptually opposed to necrosis. In view of
the inherent difficulty in accurately detecting apoptosis in chondrocytes, this chapter describes
complementary techniques that may be used in combination. During apoptotic death, protein
and DNA breakdown is accomplished by caspases (cysteine proteases) in a highly regulated
manner. Activation of caspases occurs in the initiation and/or the execution phase of certain
apoptotic programs and represents an early physiologic marker of apoptosis. Here we present
an immunoblotting technique that allows the detection of caspase-3 processing in cultured
human chondrocytes. Apoptosis leads to plasma membrane asymmetry and to externalization
of phosphatidylserine residues, which are bound with high affinity by annexin V. In the early
stages of apoptosis, cells typically have an intact cell membrane. Apoptotic cells will not stain
positive with propidium iodide, whereas externalization of phosphatidylserine will be detected
by annexin V. Terminal deoxynucleotidyl transferase (TdT)-mediated nick-end labeling
(TUNEL) works on the principle that DNA strand breaks (single or double) that occur during
apoptosis can be identified by labeling free 3'-hydroxyl termini. Labeled nucleotides are poly-
merized to these termini in a reaction catalyzed by TdT. The tissue can then be examined histo-
logically for identification of TUNEL-positive cells in situ.
Key Words: Apoptosis; caspases; DNA strand breaks; TUNEL; annexin V; cell death; chon-
drocyte death; mechanical injury; cartilage.
1. Introduction
Apoptosis is a form of programmed cell death conceptually opposed to necro-
sis. Apoptosis is a systematic breakdown of cellular structures and is less likely
to cause inflammation than necrosis. During necrosis, uncoordinated breakdown
of cellular constituents occurs by lysosomal enzymes. During apoptotic death,
protein and DNA breakdown is accomplished by caspases (cysteine proteases)
in a highly regulated manner (1). Key morphologic differences between apoptosis
275
276 D’Lima, Kuhn, and Lotz
Table 1
Key Differences Between Apoptosis and Necrosis
Apoptosis Necrosis
Plasma membrane usually intact early Plasma membrane disrupted early
No leakage of cell contents Leakage of cell contents
Minimal inflammation Inflammation
Cell shrinkage (early) Cell swelling
Chromatin condensation Poorly defined clumping of chromatin
Karyorrhexis Karyolysis
(nuclear convolution and breakdown) (nuclear dissolution and disappearance)
Apoptotic bodies (late) Cell disintegration (late)
and necrosis are listed in Table 1. However, it is not always possible to differen-
tiate clearly between apoptosis and necrosis (2). Secondary necrosis or
postapoptotic necrosis may occur associated with the late release of lysosomal
enzymes, especially in the absence of phagocytosis (3–5). Some toxins can cause
apoptosis at low doses and can cause necrosis at high doses (6). Inhibition of
apoptosis (by inhibition of caspases) can result in necrosis (7,8). Other forms of
cell death such as “dark chondrocytes” or “paralyzed cells,” which do not bear
all the hallmarks of apoptosis, have been reported (9). In view of the inherent
difficulty in accurately detecting apoptosis in chondrocytes, this chapter describes
several complementary techniques that may be used in combination.
2. Materials
2.1. Detection of Caspase-3 Processing in Cultured Human
Chondrocytes Challenged With Various Death Inducers
1. Primary human chondrocytes.
2. Glucosamine, staurosporine, and sodium nitroprusside (Sigma-Aldrich, St.
Louis, MO).
3. Agonistic anti-CD95 antibody, clone CH-11 (Immunotech, Marseille, France).
4. Proteasome inhibitor 1 (PS1; Alexis, Carlsbad, CA).
5. Linbro® 4-well culture plates (ICN, Aurora, OH).
6. Tabletop centrifuge with RTH750 rotor (Sorvall, Newtown, CT).
7. Microcentrifuge (Sorvall).
8. Gel running device for polyacrylamide gel electrophoresis (PAGE) plus blotting
apparatus (Novex, San Diego, CA).
9. Nitrocellulose transfer membrane, 0.1 mM pore size (Schleicher & Schuell,
Keene, NH).
10. Anti-active caspase-3 antibody (Imgenex, San Diego, CA).
11. Anti-GAPDH (reduced glyceraldehydes-phosphate dehydrogenase) antibody

(Ambion, Austin, TX).
12. Goat anti-mouse horseradish peroxidase conjugated antibody (Santa Cruz Bio-
technology, Santa Cruz, CA).
13. Supersignal West Pico Enhanced Chemiluminescent Substrate (Pierce, Rock-
ford, IL).
14. Biomax MR Film (Eastman-Kodak, Rochester, NY).
15. Autoradiography cassette (Fisher Scientific, Pittsburgh, PA).
2.2. Quantification of Cell Death in Glucosamine-Stimulated Cultured

Chondrocytes by Uptake of Propidium Iodide and Binding of Annexin V
1. Primary human chondrocytes.
2. Propidium iodide (Sigma-Aldrich).
3. Fluorescein-labeled Annexin V Staining Kit (BioVision, Mountain View, CA).
4. Six-well plates (Corning).
5. Centrifuge tubes (Stockwell Scientific, Scottsdale, AZ).
6. Tabletop centrifuge with RTH750 rotor (Sorvall).
7. Microcentrifuge (Sorvall).
8. 5-mL Round-bottomed fluorescence-activated cell sorting (FACS) tubes (Becton
Dickinson, Franklin Lakes, NJ).
9. Fluorescence activated cell sorter (Becton-Dickinson).
2.3. In Situ Detection of Apoptosis by TUNEL

1. A mechanical compression device (servohydraulic testing machine, cat. no. 8511;
Instron, Boston, MA).
2. Fresh (less than 72 h after death) full-thickness human articular cartilage (from
tissue banks or autopsies).
3. Glass slides (treated with silane are recommended), cover slips, mounting medium
(90% glycerin, 10% phosphate-buffered saline [PBS])
4. 4% Paraformaldehyde buffered at pH 7.4 for fixation.
5. PBS.
6. Xylene.
7. Ethanol: 100, 90, and 80%.
8. 20 µg/mL Proteinase K.
9. Terminal deoxynucleotidyl transferase (TdT; 2.5 U/mL; Roche Diagnostics, India-
napolis, IN).
10. TdT buffer: 30 mM Tris base (pH 7.2), 140 mM sodium cacodylate, 2 mM cobalt
chloride (Roche Diagnostics).
11. Fluorescein isothiocyanate (FITC)-dUTP (Roche Diagnostics; also part of the
Mebstain kit, MBL, Nagoya, Japan).
12. Termination buffer: 300 mM NaCl, 30 mM sodium citrate (Sigma-Aldrich).
13. Counterstain: propidium iodide, 0.5 µg/mL (Sigma-Aldrich).
14. DNase buffer: 100 µg/mL DNase I, 30 mM Tris base pH 7.2, 140 mM K cacody-
late, 4 mM MgCl2, 0.1 mM dithiothreitol (Roche Diagnostics).
3. Methods
3.1. Detection of Caspase-3 Processing in Cultured Human
Chondrocytes Challenged With Various Death Inducers
Caspases are zymogens that, in response to a proapoptotic input, are pro-
cessed into complex-forming large and small fragments either by other
caspases or autocatalytically. Activation of caspases occurs in the initiation
and/or the execution phase of certain apoptotic programs and represents an
early physiologic marker of apoptosis resulting, for example, in the cleavage
of substrates such as poly-ADP ribose polymerase (PARP) or lamin (10), or
the activation of caspase-activated DNase (CAD) (11). The activation of
caspases can be monitored fluorometrically by measuring the cleavage of
caspase-specific fluorogenic peptide substrates (12). Although this approach
is simple, fast, and quantitative, intercaspase or interprotease reactivity of the
substrates may generate experimental artifacts. Monitoring caspase process-
ing by immunoblotting represents a more reliable approach to assess caspase
activation. Here we present an immunoblotting technique that allows the
detection of caspase-3 processing in cultured human chondrocytes. Cell death
is induced by stimulation with staurosporine, sodium nitroprusside (SNP), glu-
cosamine, or agonistic CH-11 antibody with a proteasome inhibitor, PS1 (CH-
11/PS1).
3.1.1. Chondrocyte Isolation and Culture and Induction in Cell Death

1. Human chondrocytes were obtained from knee articular cartilage of donors
without a recorded history of joint disease. Chondrocytes were isolated from
human cartilage using established protocols. The cartilage tissue was processed
as described previously (13). The chondrocytes used in this experiment were
treated as follows:
a. After initial medium isolation, the cells were incubated in Dulbecco’s modi-
fied Eagle’s medium (DMEM) supplemented with 25 mM glucose (Gibco),
10% calf serum (Omega Scientific), 10 mM L-glutamine (Gibco), and penicil-
lin/streptomycin (Gibco) and were allowed to attach to the culture flasks.
b. After incubation the cells were grown to confluence (for approx 3 wk), split
once (passage 1), and grown to confluence again.
2. Passage 1 chondrocytes were plated at 1 × 106 cells/well (i.e., confluent) in 4-well
plates in DMEM as described above.
3. For stimulation with agonistic CH-11 antibody + proteasome inhibitor PS1
(CH-11/PS1), staurosporine, or SNP, the cells were incubated overnight, and
the medium was changed to fresh medium; for stimulation with glucosamine,
the cells were cultured overnight and the medium was changed to DMEM
supplemented with 5 mM glucose, 10 mM glutamine, penicillin/streptomycin,
and 2% calf serum. All cultures were incubated for an additional 24 h.
Fig. 1. Immunoblot analysis of caspase-3 processing. The chondrocytes were

stimulated with staurosporine (St), SNP, glucosamine (GlcN), or CH-11/PS1 (C/P)
to induce cell death. Control chondrocytes (co) were not stimulated. The effects of
CH-11/PS1 stimulation on two different donors (D1, D2) are shown. Signals for
GAPDH obtained upon reprobing of the blot suggest that similar amounts of protein
(approx 25 µg) were loaded in each lane. The large fragment of caspase-3 occurs in
three distinct processing forms of approx 22, 20, and 17 kDa.
4. The cells were then stimulated with 20 mM glucosamine, 0.5 mg/mL

staurosporine, 1 mM SNP, or CH-11/PS1 (1 mg/mL CH-11 + 20 mM PS1) for
16–24 h (see Note 1.).
5. The cells were harvested by collecting the supernatants, which contained the
floating cells, and by trypsinizing the adherent cells; supernatants and adherent
cells were combined and centrifuged for 5 min at 800g using a Sorvall tabletop
centrifuge equipped with an RTH750 rotor.
6. The supernatants were carefully and completely aspirated.
3.1.2. Immunoblotting
Whole cell lysates were prepared from approx 1–1.5 × 106 chondrocytes
that were either left unstimulated (control) or stimulated as indicated in Fig. 1.
1. The cell pellets from step 6, Subheading 3.1.1. were lysed with 0.2 mL of
ice cold lysis buffer (10 mM Tris, pH 7.6, 158 mM NaCl, 1 mM EDTA, 0.1%
sodium dodecyl sulfate [SDS], 1% Triton X-100, leupeptin, aprotinin [1 mg/mL
each], and 1 mM phenylmethylsulfonyl fluoride [PMSF; which was added imme-
diately before use]).
2. The lysates were transferred to Eppendorf tubes, and the protein concentration
was determined using a protein dye reagent from Bio-Rad according to the proto-
col supplied.
3. Similar amounts of protein were separated by SDS-PAGE for 2 h using 15%
polyacrylamide gels (Novex).
4. The proteins were transferred to nitrocellulose filters by electroblotting for 1.5 h

at 20 V (Novex blotting apparatus); the large fragment of active caspase-3 ranges
in molecular weight from 17 to 22 kDa (see Note 2.).
5. The filters were blocked overnight at 4°C in Tris-buffered saline solution supple-
mented with 0.1% Tween-20 (TBS-T) and 4% calf serum. The membrane was
then incubated with antibody against the p17 subunit of caspase-3 (1:1000) for
2 h at room temperature. The membrane was washed three times for 10 min with
TBS-T and then further incubated with horseradish peroxidase-conjugated sec-
ondary antibody (1:1000) for 1 h. Afterward the membrane was washed three
times with TBS-T and was developed using the enhanced chemiluminescent sub-
strate from Pierce.
6. The blot was exposed for 5 min to obtain optimal signals for the large fragment
of active caspase-3.
7. The blot was then normalized for protein loading: the membrane was washed
twice for 5 min with TBS-T and blocked for 1 h in TBS-T plus 4% calf serum.
8. The blot was incubated with antibody against GAPDH, 1:3000, for 1 h, washed
three times with TBS-T, and further incubated with horseradish peroxidase-con-
jugated secondary antibody (1:3000) for 1 h.
9. The membrane was washed three times, and the blot was developed using the
enhanced chemiluminescent substrate from Pierce.
10. The blot was exposed for 2 s to obtain optimal signals for GAPDH.
3.1.3. Interpretation of the Results

Despite induction of substantial cell death, neither glucosamine, SNP, nor
staurosporine induced detectable processing of procaspase-3 into its active
form (Fig. 1). In addition, glucosamine appeared to reduce the expression of
procaspase-3 in chondrocytes. Cell death induced by any one of the three
stimuli is therefore unlikely to utilize pathways that converge or depend on the
activation of caspase-3. In contrast, CH-11/PS1 stimulation causes substantial
induction of procaspase-3 processing. This stimulus may therefore serve as a
model for the induction of “classic” apoptosis in cultured chondrocytes.
3.2. Quantification of Cell Death in Glucosamine-Stimulated Cultured

Chondrocytes by Uptake of Propidium Iodide and Binding of Annexin V
The method is based on the loss of membrane integrity in dead or dying
cells, which predisposes them for the uptake of propidium iodide. Apoptosis,
as well as necrosis, leads to plasma membrane asymmetry and to exter-
nalization of phosphatidylserine residues, which are bound with high affinity
by annexin V (14,15). The technique described here is fast and simple and
allows for the simultaneous processing of a large number of samples. Unlike
methods that are based on the measurement of mitochondrial function, DNA
fragmentation, or adhesion assays, this approach directly measures cell death.
Necrosis is associated with early disruption of the cell membrane; therefore,

necrotic cells will stain positive with both propidium iodide and annexin V. In
the early stages of apoptosis, cells typically have an intact cell membrane.
Apoptotic cells will not stain positive with propidium iodide, whereas exter-
nalization of phosphatidylserine will be detected by annexin V.
3.2.1. Isolation of Chondrocytes
1. Isolate chondrocytes from human cartilage using established protocols as
described above in Subheading 3.1.1.
2. Plate chondrocytes at 50,000 cells/well (i.e., subconfluent) in 6-well plates using
3 mL of DMEM supplemented with 25 mM glucose (Gibco), 10 mM glutamine
(Gibco), penicillin/streptomycin (Gibco), and 10% calf serum (Omega Scientific).
3. Culture cells overnight and change medium to DMEM supplemented with 5 mM
glucose, 10 mM glutamine, penicillin/streptomycin, and 2% calf serum.
4. Incubate for 24 h and stimulate the cells with 5, 10, or 20 mM of glucosamine for
another 16–24 h.
5. Collect the supernatants (floating cells); trypsinize (0.5 mL of trypsin solution
per well) the adherent cells; add 3 mL of medium to trypsinized cells, and com-
bine supernatants and adherent cells in centrifuge tubes. Centrifuge for 5 min at
800g using a tabletop centrifuge.
6. Resuspend pellets in 1 mL of PBS containing 0.4% calf serum, and transfer the
cell suspensions to Eppendorf tubes. The cells may form clumps owing to the
loss of membrane integrity; a single-cell suspension can typically be obtained by
carefully forcing the cells through an 18-gage needle attached to a 1-mL syringe.
7. Centrifuge at 800g using a microcentrifuge.
3.2.2. Staining of the Cells

1. Resuspend cell pellets in 0.5 mL of Annexin V binding buffer and add 1 µL of
Annexin V-FITC supplied with the kit; mix briefly.
2. Incubate for 10 min in the dark.
3. Centrifuge at 800g.
4. Wash cells once with 0.5 mL of PBS containing 0.4% calf serum.
5. Resuspend in 0.5 mL of PBS + 0.4% calf serum and add propidium iodide (final
concentration 10 mg/mL); mix briefly.
6. Transfer the cell suspensions to FACS tubes.
7. Immediately analyze the cells by flow cytometry; Annexin V-FITC is detected
using the FL-1 channel, and propidium iodide (PI) is detected using the FL-3
channel.
8. Analyze results using CellQuest software (Becton Dickinson).

At 5 mM, as well as at 10 mM, glucosamine induces only a marginal increase
in propidium iodide uptake (compare upper left quadrants in Fig. 2). At these
Fig. 2. FACS analysis of glucosamine (GlcN)-stimulated chondrocytes. The cells

were stimulated with the indicated concentrations of glucosamine and harvested 20 h
later. Glucosamine induces a dose-dependent uptake of propidium iodide but limited
binding of annexin V, which is apparent only at 20 mM. To distinguish between cell
death modalities, a time-course analysis should be performed to determine whether
annexin V binding occurs prior to or simultaneously with propidium iodide uptake.
concentrations there is no evidence for annexin V binding (compare upper and

lower right quadrants). However, incubation with 20 mM glucosamine causes a
substantial increase in propidium iodide- as well as annexin V-positive cells
indicating that at this concentration glucosamine causes significant membrane
damage. The result suggests that all annexin V-positive cells, are also positive
for propidium iodide, indicating that annexin V binding does not occur without
membrane disintegration. The conclusion that these cells died by an apoptotic
mechanism involving phosphatidylserine externalization without cell membrane
disruption is therefore not supported.
3.3. In Situ Detection of Apoptosis by TUNEL
Several techniques exist to detect apoptosis in situ in cartilage tissue. These
include detection of DNA breaks, poly-ADP ribose polymerase (PARP) cleav-
age, phosphatidylserine exposure, and morphologic analysis. This section will
focus on the detection of DNA breaks using TUNEL (TdT-mediated nick-end
labeling). TUNEL has been one of the most common methods used to detect
apoptosis since it was described by Gavrieli et al. in 1992 (16). TUNEL works
on the principle that DNA strand breaks (single or double) can be identified by
labeling free 3'-hydroxyl termini. Labeled nucleotides are polymerized to these
termini in a reaction catalyzed by TdT. The tissue can then be examined histo-
logically for identification of TUNEL-positive cells.
Several recent reports have raised issues with regard to the specificity of
TUNEL in detecting apoptosis. Free 3'-DNA ends can occur during random
necrotic DNA degradation. TUNEL has been shown to be positive in necrotic
and autolytic cells (17–19) and also in living cells (9). However, no single test
has emerged to detect apoptosis in chondrocytes in situ with sufficient specific-
ity and sensitivity. TUNEL is a relatively efficient test that facilitates cell count-
ing and is available as commercial kits. Therefore, the use of TUNEL, combined
with morphological analysis and adequate positive and negative controls, should
yield reasonably accurate results. Loss of phospholipid asymmetry in the cell
membrane results in externalization of phosphatidylserine. As described above,
fluorescently labeled annexin V (which binds specifically to phosphatidylserine)
is one method of detecting this phenomenon in apoptosis. Recently, antibodies
specifically detecting active caspase-3 or neoepitopes in cleaved caspase sub-
strates such as the 85-kDa fragment of PARP have been used to identify
apoptotic cells in tissues including cartilage (20). These approaches may be
valuable adjunct tools for identifying apoptosis in cartilage when used in com-
bination with TUNEL.
3.3.1. Chondrocyte Death Through Cartilage Mechanical Injury

Several reports have linked chondrocyte apoptosis with mechanical injury
(21–27). A simple method of inducing cell death is presented here.
1. Obtain fresh (within 72 h of death) full-thickness human articular cartilage from
femoral condyles, tibial plateaus, or patellae.
2. Cut 5-mm diameter disks using a 5 mM dermal punch (Acuderm, Ft. Lauder-
dale, FL).
3. Culture in DMEM supplemented with L-glutamine, antibiotics, and ascorbic acid
for 48–72 h to allow the cells and tissue to stabilize.
4. Load the cartilage explants in a mechanical compression device (Instron, cat. no.
8511 servohydraulic testing machine). See Note 3 for alternative methods of
mechanical compression. A small preload should be applied for 2 min and the
cartilage thickness measured. Compress the cartilage explants (30% strain for
500 ms).
5. Sham controls should be placed in the machine and preloaded but should not be
subjected to 30% strain.
6. Remove explants and culture in fresh media for 48 h before fixation.
3.3.2. Tissue Preparation

1. Fix the tissue samples overnight in 4% buffered paraformaldehyde and embed in
paraffin. The paraformaldehyde fixes the tissue by crosslinking and prevents the
loss of DNA during the ethanol treatment.
2. Microtome tissue sections between 2 and 5 µm thick.
3. Deparaffinize the tissue sections by heating in an incubator for 30 min at 60°C
(see Note 4). Then rehydrate with xylene (three times, 3 min each), 100% ethanol
(twice, 3 min each), 90% ethanol (once, 3 min), 80% ethanol (once, 3 min), and
finally wash in distilled water (four times, 2 min each).
3.3.3. Digestion
1. Incubate in PBS at 37°C for 30 min.
2. Tap off the PBS from the slide and add 250 µL of proteinase K solution on the
section (see Notes 5–7). Incubate again at 37°C for 30 min.
3. Wash slide with distilled water (four times, 2 min each).
3.3.4. DNA Nick-End Labeling

1. Immerse (or pipet) 50 µL TdT buffer.
2. Prepare the TdT solution: 45 µL TdT buffer + 2.5 µL FITC-dUTP + 2.5 µL TdT
(see Note 8).
3. Incubate the section in TdT solution at 37°C for 1 h.
4. Halt the reaction by incubating the section in termination buffer at 37°C for 15 min.
5. Wash in PBS solution (three times, 5 min each).
3.3.5. Counterstain
1. Pipet 50 µL of propidium iodide onto the section and incubate at 4°C for 20 min.
2. Wash with PBS for 3 min.
3.3.6. Controls
1. Digest with DNase buffer (10 min at 20°C, followed by distilled water washes)
before DNA nick-end labeling. DNase I produces DNA strand breaks at the 3'-
OH end (similar to DNA fragments found in apoptosis).
2. Negative control: do not add TdT to the TdT solution. This will not catalyze the
polymerization of nucleotides at the DNA strand break.

Figure 3 displays the results of TUNEL in cartilage explants subjected to
30% compression. The TUNEL-positive cells are stained green, and the
TUNEL-negative cells are counterstained red (propidium iodide). This pro-
vides presumptive evidence of apoptosis. Further confirmation is necessary
and is provided by electron microscopic examination. Morphologic changes
characteristic of apoptosis are shown in Fig. 4. These include nuclear conden-
sation, cell shrinkage, and matrix vesicles (see Note 9).
Fig. 3. (A) Control (uninjured) cartilage showing three TUNEL-positive cells (white
arrows pointing to bright TUNEL-positive cells). The TUNEL-negative cells were
counterstained with propidium iodide (white arrowheads pointing to less bright cells).
(B) Cartilage subjected to mechanical injury showing widespread TUNEL-positive
cells (white arrows). Only a few cells were TUNEL-negative (white arrowheads).
4. Notes
1. For CH-11/PS1, staurosporine, and SNP an incubation period of 16 h was suffi-
cient to induce substantial cell death; for CH-11/PS1 stimulation, incubation for
more than 16 h is not recommended as it may result in a reduced cellular content
of processed caspase-3. With respect to glucosamine, incubation times may have
to be increased to achieve substantial induction of cell death.
Fig. 4. Electron microscopic images of chondrocytes. Original magnification

4700×, bar = 2 µm. (A) Normal chondrocytes. (B) Chondrocytes displaying morphol-
ogy of apopotosis: cell shrinkage, nuclear condensation, and formation of matrix
vesicles. N, nucleus; Cy, cytoplasm; MV, matrix vesicles. Reprinted from ref. 25 with
permission from The Journal of Bone and Joint Surgery, Inc.
2. The large fragment of active caspase-3 ranges in molecular weight from 17 to

22 kDa. The small size of these fragments requires careful adjustment of the
blotting conditions to ensure optimal signal intensity; the use of nitrocellulose
membranes with a 0.1 mM pore size is recommended since this reduces the effects
of “blotting through” the membrane.
3. The Instron servohydraulic machine that was used to injure cartilage is an

expensive piece of equipment. Less expensive mechanical compression
machines can be used. If access to a precision compression device is not avail-
able, manual compression can also be achieved by loading the cartilage explant
with a suitable weight that approximates the desired strain. The results may be
more variable than those achieved by using precision equipment, but cell death
can be induced.
4. A parafilm coverslip can be used to prevent drying during incubation. Fold up
one corner of the parafilm to facilitate removal
5. Frozen sections do not need proteinase K digestion.
6. The concentration of proteinase K depends on the type of tissue being stained. It
is recommended that proteinase K be titrated against positive controls to reduce
the number of false positives and false negatives. Use only nuclease-free pro-
teinase K because the presence of nucleases might lead to false-positive results.
7. Instead of proteinase K, a limited microwave digestion has been recommended
(28). Place the slide in a plastic jar containing 200 mL 0.1 M citrate buffer,
pH 6.0. Microwave at 350 W for 5 min (from the In Situ Cell Death Detection
Kit, Roche Diagnostics).
8. Fluorescent labels are easier to visualize and are easier to count automatically
through computerized image analysis.
9. Apoptosis and necrosis can have overlapping features. TUNEL has been reported
to be positive in necrosis. If it is essential to distinguish between apoptosis and
necrosis, additional confirmation using other tools such as morphological analy-
sis and PARP-1 cleavage are necessary.
References
1.
1 Thornberry, N. A. and Lazebnik, Y. (1998) Caspases: enemies within. Science
281, 1312–1316.
2.
2 Trump, B. F., Berezesky, I. K., Chang, S. H., and Phelps, P. C. (1997) The path-
ways of cell death: oncosis, apoptosis, and necrosis. Toxicol Pathol. 25, 82–88.
3.
3 Thompson, C. B. (1995) Apoptosis in the pathogenesis and treatment of disease.
Science 267, 1456–1462.
4.
4 Kroemer, G. (1995) The pharmacology of T cell apoptosis. Adv Immunol. 58,
211–296.
5.
5 Wertz, I. E. and Hanley, M. R. (1996) Diverse molecular provocation of pro-
grammed cell death. Trends Biochem Sci. 21, 359–364.
6.
6 Kroemer, G., Petit, P., Zamzami, N., Vayssiere, J. L., and Mignotte, B. (1995)
The biochemistry of programmed cell death. FASEB J. 9, 1277–1287.
7.
7 McCarthy, N. J., Whyte, M. K., Gilbert, C. S., and Evan, G. I. (1997) Inhibition of
Ced-3/ICE-related proteases does not prevent cell death induced by oncogenes,
DNA damage, or the Bcl-2 homologue Bak. J Cell Biol. 136, 215–227.
8.
8 Hirsch, T., Marchetti, P., Susin, S. A., et al. (1997) The apoptosis-necrosis para-
dox. Apoptogenic proteases activated after mitochondrial permeability transition
determine the mode of cell death. Oncogene 15, 1573–1581.
9.
9 Roach, H. I. and Clarke, N. M. (2000) Physiological cell death of chondrocytes in
vivo is not confined to apoptosis. New observations on the mammalian growth
plate. J. Bone Joint Surg. Br. 82, 601–613.
10.
10 Lazebnik, Y. A., Takahashi, A., Poirier, G. G., Kaufmann, S. H., and Earnshaw,
W. C. (1995) Characterization of the execution phase of apoptosis in vitro using
extracts from condemned-phase cells. J. Cell Sci. Suppl. 19, 41–49.
11.
11 Enari, M., Sakahira, H., Yokoyama, H., Okawa, K., Iwamatsu, A., and Nagata, S.
(1998) A caspase-activated DNase that degrades DNA during apoptosis, and its
inhibitor ICAD. Nature 391, 43–50.
12.
12 Gurtu, V., Kain, S. R., and Zhang, G. (1997) Fluorometric and colorimetric detec-
tion of caspase activity associated with apoptosis. Anal. Biochem. 251, 98–102.
13.
13 Kuhn, K., Hashimoto, S., and Lotz, M. (1999) Cell density modulates apoptosis in
human articular chondrocytes. J. Cell Physiol. 180, 439–447.
14.
14 Waring, P., Lambert, D., Sjaarda, A., Hurne, A., and Beaver, J. (1999) Increased
cell surface exposure of phosphatidylserine on propidium iodide negative thy-
mocytes undergoing death by necrosis. Cell Death Differ. 6, 624–637.
15.
15 Lecoeur, H., Prevost, M. C., and Gougeon, M. L. (2001) Oncosis is associated
with exposure of phosphatidylserine residues on the outside layer of the plasma
membrane: a reconsideration of the specificity of the annexin V/propidium iodide
assay. Cytometry 44, 65–72.
16.
16 Gavrieli, Y., Sherman, Y., and Ben Sasson, S. A. (1992) Identification of pro-
grammed cell death in situ via specific labeling of nuclear DNA fragmentation.
J. Cell Biol. 119, 493–501.
17.
17 Grasl-Kraupp, B., Ruttkay-Nedecky, B., Koudelka, H., Bukowska, K., Bursch,
W., and Schulte-Hermann, R. (1995) In situ detection of fragmented DNA
(TUNEL assay) fails to discriminate among apoptosis, necrosis, and autolytic cell
death: a cautionary note. Hepatology 21, 1465–1468.
18.
18 Charriaut-Marlangue, C. and Ben Ari, Y. (1995) A cautionary note on the use of
the TUNEL stain to determine apoptosis. Neuroreport 7, 61–64.
19.
19 Yasuda, M., Umemura, S., Osamura, R. Y., Kenjo, T., and Tsutsumi, Y. (1995)
Apoptotic cells in the human endometrium and placental villi: pitfalls in applying
the TUNEL method. Arch Histol. Cytol. 58, 185–190.
20.
20 Pelletier, J. P., Jovanovic, D. V., Lascau-Coman, V., et al. (2000) Selective inhibi-
tion of inducible nitric oxide synthase reduces progression of experimental osteoar-
thritis in vivo: possible link with the reduction in chondrocyte apoptosis and caspase
3 level. Arthritis Rheum. 43, 1290–1299.
21 Tew, S. R., Kwan, A. P., Hann, A., Thomson, B. M., and Archer, C. W. (2000)
21.
The reactions of articular cartilage to experimental wounding: role of apoptosis.
Arthritis Rheum. 43, 215–225.
22.
22 Loening, A. M., James, I. E., Levenston, M. E., et al. (2000) Injurious mechanical
compression of bovine articular cartilage induces chondrocyte apoptosis. Arch
Biochem Biophys. 381, 205–212.
23.
23 Chen, C. T., Burton-Wurster, N., Borden, C., Hueffer, K., Bloom, S. E., and Lust,
G. (2001) Chondrocyte necrosis and apoptosis in impact damaged articular carti-
lage. J. Orthop. Res. 19, 703–711.
24. D’Lima, D. D., Hashimoto, S., Chen, P. C., Colwell, C. W. Jr., and Lotz, M. K.
(2001) Human chondrocyte apoptosis in response to mechanical injury. Osteoar-
thritis Cartilage 9, 712–719.
25. D’Lima, D. D., Hashimoto, S., Chen, P. C., Lotz, M., and Colwell Jr, C. W. (2001)
In vitro and in vivo models of cartilage injury. J. Bone Joint Surg. Am. 83-A
(Suppl. 2), 22–24.
26.
26 Colwell, C. W., D’Lima, D. D., Hoenecke, H., et al. (2001) In vivo changes after
mechanical injury. Clin Orthop. 391(Suppl), S116–S123.
27.
27 Kim, H. T., Lo, M. Y., and Pillarisetty, R. (2002) Chondrocyte apoptosis follow-
ing intraarticular fracture in humans. Osteoarthritis Cartilage 10, 747–749.
28. Roach, H. I. and Clarke, N. M. (1999) “Cell paralysis” as an intermediate stage in
the programmed cell death of epiphyseal chondrocytes during development.
J. Bone Miner. Res. 14, 1367–1378.
MAP Kinases in Chondrocytes 291
20
Expression, Activity, and Regulation
of MAP Kinases in Cultured Chondrocytes
Jang-Soo Chun
Summary
The mitogen-activated protein (MAP) kinase family consists of extracellular signal-regu-
lated protein kinase (ERK), p38 kinase, and c-Jun N-terminal kinase (JNK) and transduces
signals from the extracellular environment to the cytoplasm and nucleus. MAP kinase signal-
ing involves a multistep kinase cascade including MAP kinase kinase kinase (MAPKKK),
MAP kinase kinase (MAPKK), and MAP kinase. The MAP kinase subtypes are constitutively
expressed in articular chondrocytes and they regulate chondrocyte function, including dif-
ferentiation, apoptosis, inflammatory responses, and activation of matrix metalloproteinases.
Therefore, imbalance or destruction of homeostasis regulating MAP kinase activity is related
to the pathogenesis of cartilage diseases such as osteoarthritis. This chapter describes methods
for measuring and modulating MAP kinase subtype activity in primary cultured articular
chondrocytes and cartilage explants.
Key Words: Chondrocyte; cartilage; osteoarthritis; cyclooxygenase; MAP kinase; ERK;

JNK; p38; IL-1β; nitric oxide; PGE2; protein kinase C; apoptosis; differentiation; inflamma-
tion; matrix metalloproteinase.
1. Introduction
The mitogen-activated protein (MAP) kinase family regulates a variety of
cellular responses by conveying information from the extracellular environ-
ment to the cytoplasm and nucleus (1–4). These enzymes occupy one level of
three-step kinase cascades that have been conserved during evolution in many
organisms (Fig. 1). Upon activation by an upstream component such as small
G proteins, MAP kinase kinase kinase (MAPKKK), a dual-specificity kinase,
phosphorylates and activates MAP kinase kinase (MAPKK), which in turn
phosphorylates threonine and tyrosine on MAP kinase, resulting in MAP
kinase activation. The mammalian extracellular signal-regulated protein kinase
291
292 Chun
Fig. 1. Mammalian MAP kinase pathways. ASK1, apoptosis signaling-regulating

kinase 1; ATF, activated transcription factor; CPLA2, cytosolic phospholipase A2;
ELK-1, transcription factor; ERK, extracellular signal-regulated protein kinase; JNK,
c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; MAPKAP-K1,
MAPK-activated protein kinase-1; MEF-2C, mouse embryonic fibroblast-2C;
MEKK1,4, MAPK/ERK kinase kinase 1; MKK, MAP kinase kinase; MLK, mixed
lineage kinases; Mnk1/2, MAPK-interacting kinase 1/2; MSK1, mitogen and stress-
activated protein kinase; PPARγ, peroxisome proliferator-activated receptor γ; SAP-1,
shrimp alkaline phosphatase stomach-associated PTP1.
(ERK) consists of ERK-1 (44 kDa) and ERK–2 (42 kDa), the p38 kinase fam-
ily consists of four members– (p38α, p38β, p38γ, and p38δ), and the c-Jun N-
terminal kinase (JNK) family consists of three members, JNK-1 (46 kDa),
JNK-2 (54 kDa) and JNK-3 (55 kDa). Among the MAP kinase subtypes, ERK
generally regulates growth and differentiation of cells, whereas JNK and p38
kinase subtypes mediate stress responses (3–5).
MAP kinase subtypes are constitutively expressed in most cell types, includ-
ing chondrocytes, and changes in MAP kinase activity affect a variety of cellular
responses. In articular chondrocytes, it has been shown that MAP kinases regu-
late differentiation, apoptosis, inflammatory responses, and activation of matrix
metalloproteinases (5–10). For example, nitric oxide (NO) from the NO donor
sodium nitroprusside (SNP) acts via MAP kinases to regulate apoptosis, dedif-
Fig. 2. Signaling pathways involved in nitric oxide (NO)-induced apoptosis, cyclo-

oxygenase-2 (COX-2) expression, and dedifferentiation in articular chondro-
cytes.ERK, eletracellular signal-regulated protein kinase; NFκB; nuclear factor κB;
PKC; protein kinase C.
ferentiation, and cyclooxygenases-2 (COX-2) expression in primary articular

chondrocytes (Fig. 2). However, this regulation can be complex since NO-
induced ERK activation induces dedifferentiation and inhibits apoptosis, whereas
NO-induced p38 kinase activation induces apoptosis and maintains differenti-
ated phenotypes. In addition to affecting MAP kinase activity, NO inhibits pro-
tein kinase C (PKC)α and -ζ activity. Inhibition of PKCα activity is owing to
reduced expression, which is independent of MAP kinase signaling, whereas
PKCζ activity is blocked as a result of p38 kinase activation. NO-induced activa-
tion of p38 kinase also activates nuclear factor (NF)-κB in a manner dependent
on MAP kinases and PKC. NO-induced NF-κB activation leads to apoptosis via
p53 and COX-2 expression, whereas dedifferentiation is independent of NF-κB
signaling.
The primary aim of this chapter is to describe methods for measuring and
modulating MAP kinase subtypes, activity in primary cultured chondrocytes.
2. Materials
1. Growth plate chondrocytes from 2-wk-old rabbit knee joint.
2. Recombinant human interleukin (IL)-1β.
3. Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine-calf serum, strep-
tomycin, and penicillin for cell culture (Gibco-BRL, Gaithersburg, MD).
294 Chun
4. Collagenase type II (381 U/mg solid; Sigma).

5. Antibodies against ERK, p38 kinase, and JNK (New England Biolabs, Beverly,
MA).
6. Antibodies against phosphorylated forms of ERK, p38 kinase, and JNK (New
England Biolabs).
7. Antibodies against phosphorylated forms of Elk, activating transcription factor-2
(ATF-2), and c-Jun (New England Biolabs).
8. Recombinant ATF-2, Elk, and c-Jun (New England Biolabs).
9. Peroxidase-conjugated secondary antibody (any supplier).
10. Rhodamine- or fluorescein-conjugated secondary antibody (any supplier).
11. Enhanced chemiluminescence reagents (ECL).
12. Protein A/G-sepharose beads (Pierce, Rockford, IL).
13. Lysis buffer A (for Western blot analysis): 50 mM Tris-HCl, pH 7.4, 150 mM
NaCl, 1% Nonidet P-40, and 0.1% sodium dodecyl sulfate (SDS), supplemented
with protease inhibitors (10 µg/mL leupeptin, 10 µg/mL pepstatin A, 10 µg/mL
aprotinin, and 1 mM 4-[2-aminoethyl] benzenesulfonyl fluoride) and phosphatase
inhibitors (1 mM NaF and 1 mM Na3VO4). Prepare fresh lysis buffer from con-
centrated stock solutions of 2 M Tris-HCl, 5 M NaCl, 10% SDS and 1000X pro-
tease and phosphatase inhibitors.
14. Lysis buffer B (for immune-complex kinase assay): 20 mM Tris-HCl, pH 7.5,
150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium
pyrophosphate, and 1 mM β-glycerophosphate, plus inhibitors of proteases and
phosphatases as in buffer A. Prepare fresh lysis buffer from concentrated stock
solutions of 2 M Tris-HCl, 5 M NaCl, 0.5 M EDTA, 80 mM EGTA, 250 mM
sodium pyrophosphate, 100 mM β-glycerophosphate, and 1000X protease and
phosphatase inhibitors.
15. Kinase assay buffer: 25 mM Tris-HCl, pH 7.5, 5 mM β-glycerophosphate, 2 mM
dithiothreitol (DTT), 0.1 mM sodium orthovanadate, and 10 mM MgCl2. Prepare
from concentrated stock solutions of 2 M Tris-HCl, 0.5 M MgCl2, 1 M DTT, and
1 M sodium orthovanadate.
16. [γ-32P]ATP.
17. Immunohistochemical staining kit (AEC chromogenic substrate-streptavidin-per-
oxidase system; DAKO, cat. no. K 0672).
18. Mammalian expression vector for dominant-negative and constitutively active
MAP kinase subtypes.
19. LipofectaminePLUS (Gibco-BRL).
20. Standard minigel apparatus.
21. Gel transfer apparatus.
22. Standard fluorescence microscope.
23. Pharmacological inhibitors such as PD98059, SB203580, SB202190, and
SP600125 can be obtained from Calbiochem (Darmstadt, Germany), Biomol
(Plymouth Meeting, PA), Cell Signaling (Beverly, MA), Promega (Madison, WI),
A.G. Scientific (San Diego, CA), Alexis Biochemicals (Montreal, Canada),
Sigma (St. Louis, MO), and other supplies.
3. Methods
The methods below detail the culturing of primary articular chondrocytes,
measurement of MAP kinase activity, and modulation of MAP kinase activities
using genetic tools and pharmacological inhibitors.
3.1. Primary Culture of Articular Chondrocytes

Articular chondrocytes can be obtained from human osteoarthritic joint carti-
lage of patients undergoing total knee arthroplasty, or from 2-wk-old or 3-mo-
old New Zealand white rabbit knee joint articular cartilage. Here we briefly
describe procedures for the isolation and culture of growth plate articular
chondrocytes (Detailed procedures for primary culture of chondrocytes are pro-
vided in Chaps. 1, 14, and 15).
1. Digest enzymatically cartilage slices from 2-wk-old rabbits for 6 h in 0.2% colla-
genase type II in DMEM, and obtain single cells by collecting the supernatant
after brief centrifugation.
2. Resuspend cells in DMEM supplemented with 10% (v/v) fetal calf serum, 50 µg/
mL streptomycin, and 50 U/mL penicillin.
3. Plate the cells on culture dishes at a density of 5 × 104 cells/cm2.
4. Change the medium every 1.5 d after seeding; cells are confluent by d 4 or 5.
5. Confluent primary cultures can be subcultured by plating cells at a density of 5 ×
104 cells/cm2 (10,11).
3.2. Assay of MAP Kinase Activity

Activation of chondrocyte MAP kinase subtypes following exposure to
extracellular stimuli such as IL-1β and tumor necrosis factor-α can be deter-
mined by detecting phosphorylated (hence activated) MAP kinase subtypes
using Western blot analysis. Kinase activity can be directly measured using
an in vitro immune complex kinase assay. Although there are many commer-
cially available kits, the methods below outline standard procedures for deter-
mining MAP kinase activity in primary articular chondrocytes treated with
IL-1β (1–5 ng/mL).
3.2.1. Assay of MAP Kinase Phosphorylation by Western Blot Analysis

MAP kinase subtypes are activated by dual phosphorylation on conserved
threonine and tyrosine residues. The development of commercially available
antibodies that specifically recognize dually phosphorylated MAP kinase sub-
types makes it possible to detect activated MAP kinase subtypes by Western
blot analysis (see Notes 1 and 2).
1. Incubate confluent primary chondrocytes (35- or 60-mm dishes) with activating
agent (e.g., 5 ng/mL IL-1β) for the desired time.
296 Chun
2. Wash cells twice with ice-cold phosphate-buffered saline (PBS), add ice-cold
lysis buffer A (150 or 300 µL for 35- or 60-mm dishes, respectively), harvest
cells by scraping, and place suspension in a 1.5-mL microcentrifuge tube.
3. Incubate on ice for 30 min with occasional vortexing, pellet cellular debris by
microcentrifugation for 10 min at 4°C (10,000g), and collect supernatant (total cell
lysate). In general, up to 150 and 300 µg of supernatant protein can be obtained
from 80–90% confluent chondrocytes on 35- and 60-mm dishes, respectively.
4. Separate proteins (30 µg/lane) by standard 8% SDS-polyacrylamide gel electro-
phoresis (SDS-PAGE) using a minigel apparatus, and transfer to a nitrocellulose
membrane.
5. After blocking the nitrocellulose sheet with 3% nonfat dry milk in Tris-buffered
saline, incubate the membrane with 1 µg/mL of primary antibody (i.e., anti-
p-ERK, anti-p-p38 kinase, or anti-p-JNK antibodies) overnight at 4°C.
6. Develop membrane using a peroxidase-conjugated secondary antibody and the
ECL system (see Fig. 3).
3.2.2. Assay of MAP Kinase Activity by In Vitro Kinase Assay

The steps below outline the procedure for immunoprecipitation of MAP
kinase subtypes and determination of their activity by immune complex kinase
assay (see Notes 3 and 4).
1. Treat subconfluent primary chondrocytes with stimulating agents (e.g., 5 ng/mL
IL-1β) for the desired time. Two 60-mm dishes per treatment will provide a suf-
ficient amount of protein (600 µg) for immunoprecipitation.
2. Wash cells twice with ice-cold PBS, add lysis buffer B, scrape cells from plate,
and transfer suspension to a 1.5-mL microcentrifuge tube. Incubate suspension
on ice for 30 min with occasional vortexing, pellet cellular debris by microcen-
trifugation for 10 min at 4°C (10,000g), and collect supernatant for immunopre-
cipitation. Using lysis buffer B, adjust protein concentration and volume to
approx 300–600 µg/600 µL.
3. Add 25 µL Protein A/Protein G-sepharose slurry and 1 µg antibody against ERK,
p38, kinase or JNK, and rock overnight at 4°C.
4. Pellet immune complexes by microcentrifugation for 2 min at 4°C (10,000g or
maximum speed in general microfuge).
5. Remove supernatant, wash pellets with lysis buffer B, and resuspend on ice in
20 µL kinase assay buffer.
6. Start kinase reactions by adding 1 µg (1 µL) of recombinant substrate (Elk for
ERK, ATF-2 for p38 kinase, and c-Jun for JNK) and 10 µCi (1 µL) [γ-32P]ATP
(see Note 3).
7. Incubate for 30 min at 30°C, then stop the reaction by adding 4X Laemmli’s
sample buffer, and boil for 3 min.
8. Resolve proteins using 10% SDS-PAGE, and detect phosphorylation of substrate
by autoradiography (see Fig. 4).
Fig. 3. Measurement of MAP kinase activity in articular chondrocytes using West-

ern blot analysis. (A) Primary articular chondrocytes were treated with 5 ng/mL
interleukin-1β (IL-1β) for the indicated times. Levels of phosphorylated extracellular
singal-regulated protein kinase (ERK)-1 and -2 were determined by Western blotting
using antibodies that specifically recognize phosphorylated ERK. The same samples
were run in parallel and used to determine ERK-2 as a control. (B) Chondrocytes were
treated with the indicated concentrations of IL-1β for 30 min. Total cell lysates were
separated by 10% SDS-PAGE, and ERK-1 was detected by Western blot analysis.
Phosphorylated ERK-1 was identified by its slower mobility. (C) Human HTB-94
chondrosarcoma cells were treated with 10 ng/mL of tumor necrosis factor-α (TNF-α)
for the indicated time periods. Phosphorylation of ERK, p38 kinase, and c-Jun
N-Terminal kinase (JNK) was determined by Western blotting using antibodies that
specifically recognize phosphorylated MAP kinase subtypes.
3.2.3. Assay of MAP Kinase Activation

by Immunohistochemistry and Immunofluorescence Microscopy
It is often not possible to obtain sufficient amounts of protein from osteoar-
thritic cartilage for biochemical analysis. The methods below outline the pro-
cedure for assaying MAP kinase subtypes in osteoarthritic cartilage or cartilage
explants by immunohistochemistry and immunofluorescence microscopy.
298 Chun
Fig. 4. Measurement of MAP kinase activity in articular chondrocytes using the

immune complex kinase assay. Primary articular chondrocytes were treated with
5 ng/mL interleukin-1β (IL-1β) for the indicated times. (A) p38α kinase activity was
determined by immunoprecipitation and in vitro kinase assay using activated tran-
scription factor-2 (ATF-2) as substrate, followed by SDS-PAGE and autoradiography.
(B) p38α kinase protein level (p38) was determined by Western blotting using a spe-
cific antibody.
3.2.3.1. IMMUNOHISTOCHEMICAL ASSAY

1. The types of samples suitable for this assay include cartilage explants (~125 mm3)
organ cultured in DMEM for 48 h in the absence or presence of 5 ng/mL IL-1β,
human osteoarthritic joint cartilage obtained from patients undergoing total knee
arthroplasty, and cartilage from experimental rheumatoid arthritic joints of
12-wk-old male DBA/1 mice (12).
2. Fix samples in 4% paraformaldehyde in PBS at 4°C overnight, and then wash
with PBS.
3. Dehydrate samples with graded ethanol (once in 30, 50, 70, 80, 90, and 95%,
and, then twice in 100%, each for 15 min), embed in paraffin, and cut 4-µm
sections using standard protocols.
4. Deparaffinate sections with xylene, and rehydrate with graded ethanol.
5. Incubate sections overnight at 4°C with antibodies (1 µg/mL) against p-ERK,
p-p38, kinase or p-JNK.
6. Wash with PBS (three times), and visualize MAP kinase by using a secondary
antibody and the AEC substrate-streptavidin-peroxidase kit (DAKO) following
the manufacturer’s instructions.
3.2.3.2. IMMUNOFLUORESCENCE MICROSCOPY

1. Expression and distribution of phosphorylated MAP kinase subtypes can also be
determined by indirect immunofluorescence microscopy using cartilage explants,
arthritic joint cartilage, and articular chondrocytes in primary culture.
2. To do this, cartilage tissue should be cryosectioned instead of paraffin embedded
for better detection of MAP kinase subtypes. Articular chondrocytes should be
cultured on round cover slips (12 mm, Bellco Glass, Vineland, NJ).
Fig. 5. Immunohistochemical detection of cyclooxygenase-2 in osteoarthritic (OA)

cartilage and cartilage explants treated with IL-1β. Cyclooxygenase-2 protein was
detected from undamaged part of (OA) cartilage (normal) and OA human joint carti-
lage (upper panels) and rabbit cartilage explants untreated (control) or treated with
5 ng/mL interleukin-1β (IL-1β) for 72 h (lower panels). Tissue sections from explants
were counterstained with hematoxylin.
3. Fix chondrocytes on cover slips (or cryosections) in 4% paraformaldehyde in PBS

containing 10% fetal calf serum at 4°C for 10 min and wash (four times) with PBS.
4. Incubate cover slip for 1 h with primary antibody (1 µg/mL). After washing (four
times) with PBS, incubate with rhodamine- or fluorescein-conjugated secondary
antibodies (2 µg/mL) for 30 min, and observe under a standard fluorescence mi-
croscope (Fig. 5).
3.3. Modulation of MAP Kinase Activation

Modulating MAP kinase activity allows for study of the role of MAP kinases
in chondrocyte functions. Described in this subheading are pharmacological
and genetic tools that can be used to modulate MAP kinase activity.
300
300
Fig. 6. Structures of pharmacological inhibitors of MAP kinases. PD98059 inhibits extracellar signal-regulated protein
kinase (ERK) activation by preventing MEK-1 activation. U0126 is a highly selective inhibitor of MEK-1 and -2. SB203580
and SB202190 inhibit p38α and p38β kinase activity. SP600125 blocks activation of c-Jun N-Terminal kinase (JNK)-1,
JNK-2, and JNK-3.
Chun
Fig. 7. Inhibition of extracellular signal-regulated protein kinase (ERK), p38

kinase, and JNK by specific chemical inhibitors. (A,B), Primary articular chon-
drocytes or (C) HTB-94 chondrosarcoma cells were treated with the indicated concen-
trations of (A) PD98059 (PD), (B) SB203580 (SB), and (C), SP600125 for 1 h and
then incubated with 5 ng/mL interleukin-1β (IL-1β) or 10 ng/mL tumor necrosis fac-
tor-α (TNF-α) for 12 h. Phosphorylation of ERK and expression of cyclooxygenase-2
(COX-2), vascular cell adhesion molecule (VCAM)-1, and intercellular adhesion mol-
ecule-1 (ICAM-1) were determined by Western blot analysis. p38 kinase activity was
determined by immune complex kinase assay. (D) Prostaglandin E2 (PGE2) synthesis
was determined by using an assay kit (Amersham Pharmacia Biotech). AFT-2, acti-
vated transcription factor-2; PD, PD98059, a MEK-1/2 inhibitor; SB, SB203580, a
p38 kinase inhibitor; SP600125, a JNK inhibitor.
3.3.1. ERK Inhibitors

1. PD98059 (Fig. 6) is a flavone compound that binds to the inactive form of MAP
kinase kinase (MEK-1), preventing its activation (13). PD98059 appears to stabi-
lize the inactive form of MEK-1, and, unlike inhibitors of other MAP kinases,
does not compete for ATP binding sites. PD98059 is 5–10-fold less effective on
MEK-2, which is 90% identical in amino acid sequence to MEK-1. It does not
block related kinases such as MEK-3, MEK-4, ERK-1, or ERK-2, nor is it known
to inhibit other kinases (13). In primary articular chondrocytes, IL-1β activates
ERK-1 and -2, and pretreatment with PD98059 prevents ERK activation. Inhibi-
tion of ERK with PD98059 partially blocks oxygenase COX-2 expression (Fig. 7A)
but completely prevents its activity (Fig. 7D).
2. U0126 is also a highly selective inhibitor of MEK-1 and -2. U0126 shows a sig-
nificantly higher affinity for MEK-1 compared with PD98059, and PD98059 and
U0126 share a common binding site (14).
302 Chun
3.3.2. p38 Kinase Inhibitors

The mammalian p38 kinase family consists of four members, p38α, p38β,
p38γ, and p38δ. SB203580 and SB202190 (Fig. 6) are the most commonly used
inhibitors of p38 kinase. They are pyridinyl imidazoles that inhibit p38α and
p38β but not other kinases such as p38γ, p38δ, ERK, JNK, or MEK (15,16).
These compounds inhibit p38 kinase by binding to the ATP binding site.
SB203580 and SB202190 are also effective in treating inflammatory disease
such as rheumatoid arthritis. IL-1β-stimulated p38α kinase activity and COX-2
expression and activity were inhibited by SB203580 pretreatment in primary
articular chondrocytes (Fig. 7B,D).
3.3.3. JNK Inhibitors

Although the mammalian JNK family consists of three members, JNK-1,
JNK-2, and JNK-3, splice variants result in a total of 10 isoforms. SP600125
(Fig. 6) (1,9-pyrazoloanthrone) is a potent, cell-permeable, reversible inhibitor
of all JNK family members. SP600125 competes with ATP binding and exhib-
its over 300-fold higher selectivity for JNK compared with ERK and p38 kinase
(17). Recently, a cell-permeable peptide inhibitor of JNK, BC-095 (Kamiya
Biomedical, Seattle, WA), has been developed, which blocks interaction of JNK
with its substrate, rather than competing with the ATP-binding site (18). In the
HTB-94 chondrosarcoma cell line, inhibition of JNK with SP600125 blocked
tumor necrosis factor-α-induced vascular cell adhesion molecule-1 expression
(Fig. 7C), whereas modulation of other MAP kinase subtypes had no effect.
3.3.4. Genetic Approaches

Although chemical inhibitors are useful for dissecting the roles of MAP ki-
nases in signaling pathways, they inevitably cause undesired side effects. The
use of dominant negative or constitutively active mutants of MAP kinase sub-
types is an alternative and possibly more physiological way to modulate MAP
kinase activities. For example, mutating Lys71 and Lys52 to Arg in ERK-1 and
-2, respectively, generates dominant negative kinase forms (19). Replacement
of Thr180 and Tyr182 with Ala and Phe, respectively, creates a dominant-nega-
tive p38α kinase, whereas, Lys55 to Arg substitution generates a dominant-
negative JNK-3 (20). MAP kinase subtypes can also be constitutively activated
by the expression of mutant upstream kinases. For example, substituting Ser189
and Thr193 for Glu in MAP kinase kinase (MKK3) and substituting Ser207 and
Thr211 for Glu in MKK6 constitutively activate p38 kinase (21).
Infection of primary chondrocytes with adenovirus or retrovirus carrying
mutant MAP kinase cDNA resulted in approx 80% of cells overexpressing
mutant. Although lipid-based transfection methods are less efficient, still more
Fig. 8. Modulation of apoptosis by dominant negative mutants of ERK and p38α

kinase. Primary culture articular chondrocytes were untreated or treated for 30 min
with 20 µM PD98059 (PD) or 20 µM SB203580 (SB). Alternatively, chondrocytes
were transfected with dominant-negative extracellular signal-regulated protein kinase
(ERK)-2 (∆ERK) or p38α kinase (∆p38) and cultured in complete medium for 24 h.
Cells were then treated for 24 h with 1 mM sodium nitroprusside (SNP), a nitric oxide
(NO) donor that causes apoptosis in chondrocytes. Apoptotic cells were determined
by FACS sorting. Treatment with PD98059 or expression of dominant-negative ERK-2
prevented NO-induced ERK activation (data not shown), whereas treatment with
SB203580 or expression of dominant-negative p38α kinase blocked NO-induced p38α
kinase activation (data not shown).
than 30% of cells can be transfected and MAP kinase activities modulated
(7–12). The method below outlines a procedure for transfecting chondrocytes
with mammalian expression vectors (see Fig. 8).
1. Seed chondrocytes at a density of 5 × 104 cells/cm2 on 35- or 60-mm dishes.
2. When cells are 50–60% confluent (d 3), transfect them with the expression vec-
tor using the Lipofectamine reagent (Gibco-BRL) following the procedure rec-
ommended by the manufacturer. We found the best transfection efficiency
in chondrocytes when using 2 µg vector, 6 µL PLUS reagent, 100 µL dilution
medium, 4 µL Lipofectamine reagent, and 0.8 mL serum-free transfection me-
dium in 1 mL transfection volume (35-mm dish).
3. Incubate at 37°C for 4 h, and then replace transfection medium with serum-con-
taining complete culture medium.
4. Culture cells for an additional 36 h prior to assaying the effects of mutant MAP
kinase overexpression.
4. Notes
1. Commercially available antibodies against phosphorylated MAP kinase subtypes
efficiently recognize human p-ERK, p-p38 kinase, and p-JNK. Of the commercial
304 Chun
antibodies raised against MAP kinases, we found that anti-p-ERK antibodies work
well in rabbit cells, but antibodies against p-p38 kinase and p-JNK are not so effec-
tive. To detect rabbit p-p38 kinase and p-JNK, it is best to use more lysate (up to
50 µg), higher antibody concentration, and longer exposure time during ECL.
2. ERK phosphorylation (i.e., activation) reduces mobility in SDS-PAGE gels.
Therefore, Western blotting using anti-ERK antibody following 10% SDS-PAGE
separation of proteins can distinguish phosphorylated and unphosphorylated ERK
(see Fig. 3B). However, it is not so easy to detect shifts in phosphorylated p38
kinase and JNK.
3. An alternative method, avoiding the use of radioisotopes, is to detect phosphory-
lation of MAP kinase substrates by Western blot analysis. To do this, the in vitro
kinase reaction is performed using “cold” ATP. Following electrophoresis and
transfer of proteins to nitrocellulose membrane, phosphorylated substrate can be
detected by Western blotting using antibodies against phosphorylated ATF-2,
Elk, or c-Jun.
4. Here we described, the use of Elk, ATF-2, and c-Jun as substrates in immune
complex kinase assays. There are other substrates for each MAP kinase subtype
such as myelin basic protein, which can be phosphorylated by each subtype.
Acknowledgments
This work was supported by the National Research Laboratory Program
(M1-0104-00-0064) of the Korea Ministry of Science and Technology.
References
1 Johnson, G. L. and Lapadat, R. (2002) Mitogen-activated protein kinase path-
1.
ways mediated by ERK, JNK, and p38 protein kinases. Science 298, 1911–1912.
2
2. Schaeffer, H. J. and Weber, M. J. (1999) Mitogen-activated protein kinases: spe-
cific messages from ubiquitous messengers. Mol. Cell. Biol. 19, 2435–2444.
3 Garrington, T. P. and Johnson, G. L. (1999) Organization and regulation of mi-
3.
togen-activated protein kinase signaling pathways. Curr. Opin. Cell. Biol. 11,
211–218.
4 Weston, C. R. and Davis R. J. (2002) The JNK signal transduction pathway. Curr.
4.
Opin. Genet. Dev. 12, 14–21.
5 Kim, S.-J., Ju, J.-W., Oh, C.-D., et al.(2002) ERK-1/2 and p38 kinase oppositely
5.
regulate nitric oxide-induced apoptosis of chondrocytes in association with p53,
caspase-3, and differentiation status. J. Biol. Chem. 277, 1332–1339.
6. Kim, S.-J., Hwang, S.-G., Shin, D.-Y., Kang, S.-S., and Chun, J.-S. (2002) p38
kinase regulates nitric oxide-induced apoptosis of articular chondrocytes by accu-
mulating p53 via NFκ B-dependent transcription and stabilization by serine 15
phosphorylation. J. Biol. Chem. 277, 33,501—33,508.
7. Kim, S.-J., Kim, H.-G., Oh, C.-D., et al.(2002) p38 kinase-dependent and -inde-
pendent inhibition of protein kinase C-ζ and -α regulates nitric oxide-induced
apoptosis and dedifferentiation of articular chondrocytes. J. Biol. Chem. 277,
30,375—30,381.
8.
8 Yoon, Y.-M., Kim, S.-J., Oh, C.-D., et al. (2002) Maintenance of differentiated
phenotype of articular chondrocytes by protein kinase C and extracellular signal-
regulated protein kinase. J. Biol. Chem. 277, 8412–8420.
9.
9 Kim, S.-J. and Chun, J.-S. (2003) Protein kinase Cα and ζ regulate nitric oxide-
induced NFκB activation that mediates cyclooxygenase-2 expression and
apoptosis but not dedifferentiation in articular chondrocytes. Biochem. Biophys.
Res. Commun. 303, 206–211.
10.
10 Yoon, Y.-M., Kim, S.-J., Oh, C.-D., et al. (2002) Maintenance of differentiated
phenotype of articular chondrocytes by protein kinase C and extracellular signal-
regulated protein kinase. J. Biol. Chem. 277, 8412–8420.
11.
11 Ryu, J.-H., Kim, S.-J., Kim, S.-H., et al. (2002) β-Catenin regulation of the chon-
drocyte phenotypes. Development 129, 5541–5550.
12 Kim, S.-J., Lim, D.-S., Kim, S.-H., et al. (2002) β-Catenin regulates expression of
12.
cyclooxygenase-2 in articular chondrocytes. Biochem. Biophys. Res. Commun.
296, 221–226.
13. Alessi, D. R., Cuenda, A., Cohen, P., Dudley, D. T., and Saltiel A. R. (1995)
PD098059 is a specific inhibitor of the activation of mitogen-activated protein
kinase kinase in vitro and in vivo. J. Biol. Chem. 270, 27,489–27,494.
14.
14 Favata, M. F., Horiuchi, K. Y., Manos, E. J., et al. (1998) Identification of a
novel inhibitor of mitogen-activated protein kinase kinase. J. Biol. Chem. 273,
18,623–18,632.
15. Jiang, Y., Chen, C., Li, Z., et al. (1996) Characterization of the structure and
function of a new mitogen-activated protein kinase (p38β). J. Biol. Chem. 271,
17,920–17,926.
16.
16 Cuenda, A., Rouse, J., Doza, Y. N., et al. (1995) SB 203580 is a specific inhibitor
of a MAP kinase homologue which is stimulated by cellular stresses and
interleukin-1. FEBS Lett. 364, 229–233.
17. Bennett, B. L., Sasaki, D. T., Murray, B. W., et al. (2001) SP600125, an
anthrapyrazolone inhibitor of Jun N-terminal kinase. Proc. Natl. Acad. Sci. USA
98, 13,681–13,686.
18.
18 Bonny, C., Oberson, A., Negri, S., Sauser, C., and Schorderet, D. F. (2001) Cell-
permeable peptide inhibitors of JNK: novel blockers of beta-cell death. Diabetes
50, 77–82.
19.
19 Robbins, D. J., Zhen, E., Owaki, H., et al. (1993) Regulation and properties of
extracellular signal-regulated protein kinases 1 and 2 in vitro. J. Biol. Chem. 268,
5097–5106.
20.
20 Raingeaud, J., Gupta, S., Rogers, J. S., et al. (1995) Pro-inflammatory cytokines and
environmental stress cause p38 mitogen-activated protein kinase activation by dual
phosphorylation on tyrosine and threonine. J. Biol. Chem. 270, 7420–7426.
21. Raingeaud, J., Whitmarsh, A. J., Barrett, T., Derijard, B., and Davis, R. J. (1996)
MKK3- and MKK6-regulated gene expression is mediated by the p38 mitogen-
activated protein kinase signal transduction pathway. Mol. Cell. Biol. 16,
1247–1255.
Chondrocyte Response to Compressive Strain 307
21
Mechanical Loading of Chondrocytes
Embedded in 3D Constructs
In Vitro Methods for Assessment of Morphological
and Metabolic Response to Compressive Strain
David A. Lee and Martin M. Knight
Summary
Mechanical loading of chondrocytes in 3D constructs has been used to investigate
mechanotransduction and its potential for stimulating tissue-engineered cartilage repair. This
chapter describes the preparation of 3D agarose or alginate constructs seeded with isolated
chondrocytes and specific test rigs for applying gross compressive strain to individual con-
structs on a confocal microscope or for longer term compression of constructs cultured within
an incubator. Experimental methods are described to quantify the level of cell deformation
and the elaboration of extracellular matrix. The chapter thus provides an introduction to the
experimental techniques used to examine chondrocyte mechanotransduction and downstream
cell function.
Key Words: Agarose; bioreactor; cartilage; cell deformation; cell mechanics; cell pro-
liferation; chondrocyte; confocal microscopy; extracellular matrix; mechanical compres-
sion; mechanotransduction; proteoglycan synthesis; sulphate incorporation; thymidine
incorporation.
1. Introduction
During normal activity, the articular cartilage of the major load-bearing
joints is exposed to a complex array of compressive, tensile, and shear forces.
It is well known that this mechanical environment is essential for the health
and homeostasis of the tissue. Previous studies have reported that whereas static
compression of cartilage causes a downregulation of matrix synthesis, cyclic
compression causes a frequency-dependent upregulation (1,2). Consequently
it has been suggested that mechanical conditioning may be applied within a
307
308 Lee and Knight
tissue engineering context to stimulate in vitro biosynthesis by cells within

three-dimensional (3D) scaffolds prior to implantation (3–5). However, the
intracellular mechanotransduction pathways through which chondrocytes are
able to sense and respond to mechanical loading remain unclear both for
chondrocytes in cartilage matrix and for isolated cells in 3D tissue-engineered
scaffolds. Potential primary signaling events include cell deformation, fluid
flow, hydrostatic pressure, and factors associated with compression of a
charged extracellular matrix such as electrical streaming potentials and changes
in pH and osmolarity (6–8).
Numerous previous studies have examined chondrocyte mechanotrans-
duction using a well-characterized experimental system consisting of isolated
chondrocytes embedded within agarose hydrogel. This in vitro model system
maintains chondrocytic phenotype over extended culture periods (9–11) while
enabling the application of cell deformation through gross compression of the
cell-agarose construct (12,13). The model system therefore enables the role of
cell deformation in cartilage mechanotransduction to be examined in isolation
from factors associated with compression of the charged extracellular matrix
(4,14–17). In addition, the model provides insight into mechanotransduction
within tissue-engineered constructs, with results indicating important differ-
ences between the potential signaling pathways within cartilage and cell-seeded
scaffolds (13).
Although compression of cell-agarose constructs will initiate transient
changes in hydrostatic pressure and fluid flow, which have both been impli-
cated in mechanotransduction, it is widely believed that cell deformation is the
primary mediator. Therefore, before examining the effect of mechanical load-
ing on chondrocyte metabolism, it is important first to identify the effect of
loading on cell morphology and deformation. In the vast majority of studies,
physiological compressive strain rather than stress is applied to the cell-agar-
ose construct with typical strains in the order of 5 to 20% (18). Various factors
have been shown to influence the level of cell deformation and hence the abil-
ity of the cells to respond to gross compressive strain. These include the pres-
ence and mechanical properties of elaborated extracellular matrix (19,20), the
modulus of the scaffold relative to that of the cells and matrix (21,22), and the
loading regime and viscoelastic properties of the scaffold (12). Having esta-
blished the nature of the cell deformation induced by mechanical loading, stud-
ies can subsequently examine the metabolic response and the associated
intracellular signalling pathways.
This chapter describes the experimental techniques employed for preparing
chondrocyte-agarose constructs, quantifying the level of cell deformation, and
analyzing the influence of compression on cell proliferation and proteoglycan
synthesis using [3H]thymidine and 35SO4, incorporation, respectively.
2. Materials
2.1. Isolation of Bovine Chondrocytes
1. Front feet from approx 18-mo-old steers. (Obtain from local slaughterhouse.)
2. 70% (v/v) Industrial methylated spirits (IMS) in water.
3. Dissection kit: metal dissection tray, and sterile scalpels fitted with nos. 20, 11,
and 15 blades.
4. Earle’s balanced salt solution (EBSS; Sigma, cat. no. E-2888 or equivalent).
5. Culture medium: Dulbecco’s minimal essential medium (Sigma, cat. no. D-5921
or equivalent) supplemented with 20% (v/v) fetal calf serum (Sigma, cat. no.
F-9665 or equivalent), 2 mM L-glutamine (200 mM stock solution, Sigma, cat. no.
G-7513 or equivalent), 20 mM HEPES (1M stock solution, Sigma, cat. no. H-0887
or equivalent), 50 U/mL penicillin plus 50 µg/mL streptomycin (100X concen-
trated stock solution, Sigma, cat. no. P0906), and 150 µg/mL L-ascorbic acid
(Sigma, cat. no. A-0278). Sterilized by filtration through a 0.2-µm-pore size filter.
Aliquot and store at –20°C prior to use.
6. Pronase solution: 700 U/mL pronase (Merck, cat. no. 39052 2P) in culture medium.
7. Collagenase solution: 100 U/mL collagenase type XI (Sigma, cat. no. C9407) in
culture medium.
8. Trypan blue solution: 0.4% (w/v) trypan blue in 0.81% (w/v) sodium chloride/
0.06% (w/v) potassium phosphate (Sigma, cat. no. T-8154 or equivalent).
9. 35-mm-Diameter sterile Petri dishes.
10. 50-mL Capacity sterile polypropylene centrifuges tubes.
11. 70-µm Pore size cell sieve (Falcon, cat. no. 35 2350).
12. Class II safety cabinet.
13. Roller mixer within an incubator or warm room at 37°C.
14. Centrifuge: up to 2000g, compatible with 50-mL centrifuge tubes.
15. Hemacytometer.
16. Inverted microscope with phase contrast facility.
2.2. Preparation of Chondrocyte/Agarose Constructs

1. Culture medium.
2. Isolated chondrocytes.
3. Agarose suspension: 6% (w/v) agarose (type VII, Sigma, cat. no. A4018,) in
EBSS, sterilized by autoclaving.
4. Mold for preparing constructs (see Note 1).
5. Positive displacement pipet and sterile tips (Gilson Microman M250 or equivalent).
6. Water bath, maintained at 37°C.
7. Roller mixer within an incubator or warm room at 37°C.
2.3. Measurement of Cell Deformation

in Compressed Cell-Seeded Scaffolds
1. Chondrocyte/agarose constructs seeded with viable cells and prepared as either
rectangular blocks or half-cores.
310 Lee and Knight
Fig. 1. Schematic diagram of the compression rig that mounts on the stage of an
inverted microscope.
2. Microscope-mounted compression rig in which compressive strain may be applied

perpendicular to the microscope axis (Fig. 1) (12,19,20,22,23).
3. Inverted microscope associated with a confocal microscope system.
4. EBSS buffer solution: EBSS plus 1mM Ca2+, 20 mM HEPES, pH 7.4.
5. Calcein-AM (Molecular Probes) viable cell staining solution prepared in culture
medium at 5 µM.
6. 5-mL Universal tube.
7. Pipets.
2.4. Compressive Cell Strain Bioreactor

The equipment comprises the following components, as illustrated in Fig. 2:
1. Loading frame to incorporate Hereaus tissue culture incubator (Zwick Testing
Machines, Leominster, UK).
2. Servo-hydraulic loading actuator incorporating load cell and Linear Variable
Displacement Transducer (Zwick Testing Machines).
3. Hydraulic power supply (Zwick Testing Machines).
4. Digital control unit (Dartec 9610, Zwick Testing Machines).
5. Hereaus B6060 incubator.
6. Custom-designed construct loading assembly and box (Department of Engineer-
ing, Queen Mary, University of London, London, UK).
For culture within the bioreactor the following are required:
1. 70% IMS.
2. Culture medium.
3. 24-Well cell culture plate (Costar, cat. no. 3524).
Chondrocyte Response to Compressive Strain
311
Fig. 2. Photograph (A) and schematic representation (B) of the compressive cell strain bioreactor. (C) (following page) Sche-
matic representation of the system during setup and in operation.
311
312 Lee and Knight
312
4. Cell seeded constructs.

5. 1-mL Capacity sterile syringe with right-angled needle (see Note 2).
2.5. Analysis of Proteoglycan Synthesis and Cell Proliferation

2.5.1. Metabolic Labeling
1. Radiolabeled medium: culture medium supplemented with 1 µCi/mL [3H]Thymi-
dine (Amersham Life Sciences, cat. no. TRK61) + 10 µCi/mL 35SO4 (Amersham
Life Sciences, cat. no. SJS1).
2.5.2. Digestion of Constructs

1. Papain digest buffer: 150 mM sodium chloride, 55 mM trisodium citrate (BDH,
cat. no. 10242), 5 mM EDTA disodium salt, 5 mM L-cysteine hydrochloride (Sigma,
cat. no. C7880) in distilled water.
2. Papain suspension (Sigma, cat. no. P3125).
3. Agarase (Sigma, cat. no. A6306).
4. Incubators set at 70°C and 37°C.
2.5.3. 35SO4 and [3H]Thymidine Incorporation

for Proteoglycan Synthesis and Cell Proliferation
1. Acetate buffer: 50 mM sodium acetate, pH 5.8, 85 mM magnesium chloride in
distilled water.
2. Alcian blue solution: 0.2% (w/v) alcian blue 8GX (Sigma, cat. no. A5268) in
acetate buffer.
3. Acetate buffer with 100 mM sodium sulfate (BDH, cat. no. 103984T).
4. Guanidine solution: 4 M guanidine HCl (Sigma, cat. no. G4630), 4.3 M propan-
2-ol (Sigma, cat. no. I9516) in distilled water.
5. Trichloroacetic acid (TCA, Sigma, cat. no. 490-10), 10% and 20% (w/v) solu-
tions in distilled water.
6. Vacuum filtration system (Millipore Multiscreen®, or equivalent), including
vacuum manifold and multiple punch.
7. Vacuum filtration plates, 0.65-µm-pore size (Millipore Multiscreen, cat. no.
MADVN6550).
8. Scintillation cocktail (Emulsifier Safe, Packard or equivalent).
9. Scintillation vials, 4-mL capacity (Packard or equivalent).
10. Liquid scintillation counter (e.g., Wallac 1409, Perkin Elmer, or equivalent).
2.5.4. DNA Analysis

1. Hoechst 33258 solution: 1 µg/mL Hoechst 33258 (Sigma, cat. no. B2883) in
150 mM sodium chloride (BDH, cat. no. 10241), 55 mM trisodium citrate
(BDH, cat. no. 10242, Poole, UK), pH 7.0.
2. DNA solutions: Calf thymus DNA (0-100 µg/mL, Sigma, cat. no. D3664) in
150 mM sodium chloride, 55 mM trisodium citrate (BDH, cat. no. 10242), pH 7.0.
3. White microtiter plates (Nunc, cat. no. 437796).
4. Microplate fluorimeter (e.g., BMG FLUOstar Galaxy, BMG Labtechnologies).
314 Lee and Knight
3. Methods
3.1. Isolation of Bovine Chondrocytes
1. Clean bovine front feet with water and immerse in 70% IMS for 5 min to sterilize
skin.
2. Within the class II safety cabinet, expose the metacarpalphalangeal joint and
remove full-depth cartilage slices from the proximal articular surfaces of the
joint using a no. 15 blade (see Note 3). Transfer the slices to a 60-mm Petri dish
containing 8 mL of EBSS.
3. Aspirate EBSS, dice the tissue into pieces approx 1 mm3, and transfer to a 50-mL
centrifuge tube containing 10 mL of pronase solution (see Note 4). Incubate on
rollers for 1 h at 37°C.
4. Allow the tissue pieces to settle and remove pronase solution. Add 30 mL of
collagenase solution and incubate on rollers for 16 h at 37°C (see Notes 4 and 5).
5. Allow any remaining tissue pieces to settle and carefully remove the collagenase
solution containing isolated cells. Filter the solution through a 70-µm pore sieve
into a fresh 50 mL centrifuge tube.
6. Centrifuge at 2000g for 5 min to pellet cells. Aspirate the supernatant and resus-
pend the cells in 10 mL of culture medium using a 10-mL sterile graduated pipet.
Repeat this process twice.
7. Add 50 µL of cell suspension to 50 µL trypan blue solution, mix, and add result-
ant suspension to a hemacytometer. Determine total cell number and cell viabil-
ity and adjust cell concentration to 2 × 107 cells/mL.
3.2. Preparation of Chondrocyte/Agarose Constructs

1. Assemble the parts of the mold used for preparation of agarose/chondrocyte con-
structs (see Note 1).
2. Add the cell suspension, adjusted to 2 × 107 cells/mL to an equal volume of
agarose suspension. Mix thoroughly using a positive displacement pipet.
3. Aliquot the agarose/chondrocyte suspension into individual wells of the mold
and complete the assembly by adding the second glass slide and clamping the
three components of the mold to ensure it is sealed.
4. Incubate for 30 min at 4°C.
5. Carefully dismantle the mold to allow removal of the gelled agarose/chondrocyte
constructs.
3.3. Measurement of Cell Deformation

in Compressed Cell-Seeded Scaffolds
1. Incubate constructs in Calcein-AM solution for 1 h at 37ºC. Allow sufficient stain-
ing solution to cover the constructs (see Note 6).
2. Transfer constructs to EBSS at room temperature to allow complete de-esterization
of Calcein-AM.
3. Place an individual construct on the glass coverslip within the compression rig, as
shown in Fig. 1.
4. Advance the compression platens until they are just touching the construct (see
Note 7).
5. Use the pipet to add EBSS buffer carefully to maintain hydration of the construct
throughout testing. Repeat as necessary.
6. Configure the confocal microscope for high-resolution imaging (pixel size 0.2
µm) with a high-magnification objective lens ( 40×). Use a laser excitation of
488 nm with fluorescent emissions detected above 500 nm.
7. To minimize boundary conditions, select individual cells that are at least 30 µm
(approx three cell diameters) from each other and from the bottom surface of the
construct. Also select cells away from the ends and sides of the construct (see
Note 8).
8. To image an individual cell, adjust the Z focus until the confocal X-Y section
image bisects the center of the cell (see Note 9).
9. In some instances it may be necessary to define the boundary of the cell more
clearly. This may be achieved by using a high-pass filter with an intensity thresh-
old. For example, previous studies have used a threshold of 50% of the maximum
intensity (19,24).
The subsequent method protocol will depend on whether a paired or non-
paired approach is employed.
3.3.1. Analysis Using a Paired Data Approach

This approach relies on the ability to image an individual cell first in the
unstrained state and then to follow that cell as gross compression is applied to the
construct, enabling the same cell to be imaged again in the compressed state.
1. Select a group of approx 10 individual cells and make a confocal section image
through the center of each cell as described above in steps 6–9, Subheading 3.3.
2. Apply the compressive strain while visualizing the cells using transmitted light
microscopy. As the compression is applied, track the individual cells by adjust-
ing the focus and the stage position as necessary. The strain rates should be less
than 5%/s (see Note 10). An example set of images is provided in Fig. 3.
3. Image the same sample of individual cells. It is important to be aware of the
viscoelastic stress relaxation properties of the scaffold since each cell is imaged
at a different time post compression. For scaffolds with an equilibrium modulus
significantly greater than that of the cells (i.e., Econstruct > 10 kPa), any relaxation
in the scaffold should not influence the level of cell deformation (22).
4. Repeat the test using a new sample on each occasion to avoid artifacts associated
with hysteresis and strain history of the scaffold.
5. Calculate the percentage change in cell X and Y diameters measured parallel and
perpendicular to the axis of compression (Fig. 4).
6. Calculate the percentage change in cell volume (π/6 · X · Y2), based on the as-
sumption that the cells deform from a spheroid to an oblate ellipsoid (Y diameter
= Z diameter). The paired data approach may also be used to measure the defor-
mation of intracellular organelles such as the nucleus (Fig. 5). Specific viable
316 Lee and Knight
Fig. 3. Confocal section images through the center of an individual cell stained
with calcein-AM and visualized within an agarose construct at 0, 5, 10, 15, 20, and
25% compressive strain. An intensity threshold has been applied to each image to
clarify the cell boundary to allow measurement of cell diameters. Scale bar = 2 µm.
Fig. 4. Percentage change in cell diameter, parallel (X) and perpendicular (Y) to the
axis of compression for freshly isolated chondrocytes compressed in 3% agarose gel.
Error bars indicate standard deviation for n = 12.
Fig. 5. Confocal images bisecting the center of a single cell (A,B) and its nucleus
(C,D) in unstrained (A,C) and 20% compressed (B,D) 3% alginate construct. The
cells and nuclei have been stained with Calcein-AM and Hoescht, respectively.
fluorescent stains should be used in conjunction with the appropriate excitation

and emission settings on the confocal microscope.
3.3.2. Analysis Using a Nonpaired Data Approach

When it is not possible to image the same individual cells in both an
unstrained and compressed state, an alternative approach may be used to esti-
mate the level of cell deformation. This so-called nonpaired data or sample
population approach uses separate samples of cells imaged in the unstrained
and compressed states. However, by using this approach direct, accurate calcu-
lation of the percentage change in cell diameter or in cell strains from the sample
populations is not possible. This is because of the large variability in unstrained
cell diameter associated with cartilage heterogeneity, such that cells isolated
from the deep zone are larger than those isolated from the surface zone (15).
318 Lee and Knight
Consequently cell deformation is estimated from individual images of cells,

using a deformation index (I).
1. Image a sample population of cells in the unstrained construct (see Note 11).
2. Apply a static compressive strain to the construct via the platens (see Note 10).
3. Allow a period of stress relaxation for the compressive stress in the gel to reach
equilibrium (see Note 12).
4. Image a new sample population of cells.
5. Repeat the test using a new sample on each occasion to avoid artifacts associated
with hysteresis and strain history of the scaffold.
6. Calculate the deformation index (I) for each cell such that I = X/Y, where X and Y
are the cell diameters measured parallel and perpendicular to the axis of com-
pression.
7. Freshly isolated chondrocytes in unstrained agarose have been shown to be
spherical and to deform at constant volume with rotational symmetry (Y = Z)
(13). Therefore, the cell diameter strains in the X and Y axes (εX and εY, respec-
tively) may be calculated from the deformation index using the following equa-
tions (22):
23
Xs
εX = – 1 × 100% (1)
Ys
–1
3
Xs
εY = – 1 × 100% (2)
Ys
3.4. Culture of Constructs in Compressive Cell Strain Bioreactor

1. Assemble loading plate, loading pins, and Perspex box, ensuring that the outer
pins are held in place with plastic collars and the inner pins are clamped (see
Fig. 2B,C). Sterilize in 70% IMS and allow to dry overnight in a tissue culture
hood.
2. Locate constructs at the center of individual wells of a 24-well cell culture
plate and locate within the Perspex box.
3. Allow loading pins to drop onto constructs by removing the plastic collars from
the outer pins and unclamping the inner pins. Clamp inner pins once in position
(see Fig. 2C).
4. Introduce 1 mL of culture medium into each well through the medium ports
associated with the apparatus using a 1-mL syringe and a right-angled needle.
5. Close the box, transfer to the bioreactor, and attach to the loading actuator by
means of the central rod. Unlock the central rod to allow free movement.
6. Set the control electronics of the straining apparatus to provide a crosshead
movement equivalent to a compressive strain ranging from 0 to 15% in a sinu-
soidal waveform at a frequency of 1 Hz.
3.5. Analysis of Cell Proliferation and Proteoglycan Synthesis

3.5.1. Radiolabeling and Digestion of Constructs (see Note 13)
1. Incubate constructs in 1 mL of radiolabeled medium at 37°C in 5% CO2 atmo-
sphere for a selected labeling period, typically between 6 and 48 h (see Note 14).
2. At the end of the labeling period, remove the medium and store at –20°C for
analysis (see Subheading 3.5.2.).
3. Place construct in 1 mL of papain digest buffer and incubate at 70°C until the
construct melts (usually 20–30 min).
4. Cool to 37°C and add 10 U/mL agarase and 1 U/mL papain. Incubate at 37°C
overnight (see Note 15).
5. Digested constructs may be stored at –20°C prior to analysis.
3.5.2. 35SO4 Incorporation for Proteoglycan Synthesis (see Note 13)

1. Add 75 µL of acetate buffer, 25 µL of digested sample or radiolabeled medium,
and 150 µL of alcian blue solution, in duplicate, into individual wells of a vacuum
filtration plate.
2. Seal the plate and incubate with agitation for 1 h at room temperature.
3. Vacuum aspirate, add 200 µL of acetate buffer with 100 mM sodium sulphate,
and incubate at room temperature for 5 min. Repeat this process twice.
4. Vacuum aspirate, carefully blot the underside of the plate to remove excess fluid,
and remove the underdrain. Place the plate in an oven, set at 37°C for 1 h.
5. Punch out the filters into individual scintillation vials using the multiple punch
and add 500 µL of guanidine solution. Incubate on a rolling mixer for 2 h to
resolubilize the precipitate.
6. Add 4 mL of scintillation cocktail and read using a scintillation counter (bound
counts in medium or construct).
7. Add 10 µL of recovered radiolabeled medium into individual scintillation vials,
add 4 mL of scintillation cocktail, and read using a scintillation counter (free
counts in medium).
8. The incorporation of sulfate may be determined as follows:
Sulfate incorpration ×
= total bound counts in medium and construct 0.81 (3)
–1
µmol SO 4⋅h ⋅µg –1 total free counts in medium × t × DNA
DNA
where t = time of culture in labeled medium in hours and 0.81 = concentration of

35SO
4 in culture medium in mM.
3.5.3. [3H]Thymidine Incorporation for Cell Proliferation (see Note 13)

1. Aliquot 200 µL of 10% TCA into individual wells of a vacuum filtration plate
and vacuum aspirate to leave the filters wet.
2. Aliquot 100 µL of 20% TCA, followed by 100 µL of digested sample, in dupli-
cate, into individual wells of the plate in duplicate.
320 Lee and Knight
3. Seal the plate and incubate with agitation for 1 h at room temperature.
4. Vacuum aspirate, add 100 µL of 10% TCA solution, and incubate at room tem-
perature for 5 min. Repeat this process twice.
5. Vacuum aspirate, carefully blot the underside of the plate to remove excess fluid
and remove the underdrain. Place the plate in an oven, set at 37°C for 1 h.
6. Punch out the filters into individual scintillation vials using the multiple punch
and add 500 µL of 0.01 M KOH. Incubate on a rolling mixer for 2 h to resolubilize
the precipitate.
7. Add 4 mL of scintillation cocktail and read using a scintillation counter.
8. [3H]Thymidine incorporation data are presented as counts per minute (counts/
min), normalized to DNA values, i.e., counts/min/µg DNA.
3.5.4. DNA Analysis

1. Aliquot 100 µL of DNA solutions, ranging from 0 to 100 µg/mL, in duplicate,
into individual wells of a white microtiter plate.
2. Add 100 µL of digested samples, in duplicate, into individual wells of a white
microtiter plate.
3. Add 100 µL of Hoechst 33258 solution into each well (see Note 16).
4. Read the plate using a microplate fluorimeter with excitation set at 360 nm and
emission at 460 nm.
5. Calculate sample DNA concentrations from the standard curve.
4. Notes
1. The molds consist of two glass slides, which act as outer supports for a medical-
grade stainless steel central unit, which incorporates a matrix of cylindrical (5-mm
diameter × 5mm depth) or rectangular holes (4 × 3 × 3 mm). All three units may be
held together using appropriate tape or clamps. For use, the lower glass slide and
central unit are clamped together and the resultant wells are filled with the chon-
drocyte/agarose suspension, taking care to avoid the formation of air bubbles. A
positive displacement pipet is ideal for this purpose, and it is recommended that
the wells be slightly overfilled at this stage. The top glass slide is carefully located
and clamped. Once gelling has occurred, the mold may be disassembled to release
cylindrical or rectangular constructs, with dimensions defined by the holes in the
central unit. The constructs have plane parallel top and bottom surfaces, which is
necessary for mechanical testing. Detailed design drawings may be obtained from
David Lee (D.A.Lee@qmul.ac.uk).
2. The right-angled needle is prepared by taking a sterile 21-gage, 50-mm-long length
needle and bending it through 90° after partially removing the needle from its
sheath.
3. Opening the joint is best performed by making a midline incision using a no. 20
blade, taking care not to enter the joint space. The skin and digital extensor ten-
dons may then be dissected away from the joint capsule. The joint may then be
opened with a no. 11 blade, using the distal surface of the metacarpal bone as a
guide.
4. The volumes of pronase and collagenase solution are based on the digestion of
tissue from one joint. If one is digesting tissue from more than one joint, it is
recommended that the tissue be digested in further centrifuge tubes rather than
altering the volume of the solutions in each tube. Approximately 1–3 g of tissue
can be obtained from a single joint, and this should yield up to 30 × 106 cells.
5. Collagenase batches vary, even when they are purchased from the same supplier.
It is highly advisable to test batches prior to use to ensure that all the tissue is
digested and that cell viability is maintained. It may be necessary to alter the con-
centration or incubation times from those stated to achieve optimal cell isolation.
6. These techniques can be combined with a viability assay by incorporating 5 µM
Ethidium Homodimer (Molecular Probes) into the staining solution. The use of
Calcein-AM ensures that cell deformation is only measured in viable cells.
7. It is useful to visualize the construct and platens using transmitted light micros-
copy to stop the platens the moment they touch the specimen.
8. At a separation of more than 30 µm between adjacent cells or the construct sur-
face, the mechanical perturbation has been shown to be less than 1% (25).
9. The position of the confocal plane bisecting the center of the cell may be selected
from an X-Z confocal scan or by taking a confocal Z series and selecting the
image in which the cell has the largest diameter (26).
10. The magnitude of the compressive strain that can be applied will depend on the
mechanical properties of the scaffold material. However, it should be noted that
at compressive strains of 25% and above cells show nonreversible membrane
blebbing around the equator of the oblate ellipsoid (26).
11. For a sample size of n = 50, the mean precision of deformation index measure-
ment in compressed agarose constructs is 0.01 at the 90% confidence interval
(26). There is minimal improvement in precision above n = 50, although this will
depend on the variability within different types of construct.
12. The length of the relaxation period will depend on the modulus and viscoelastic
properties of the construct, which may be determined from a stress relaxation test
in uniaxial unconfined compression. For constructs with an equilibrium modulus
significantly greater than that of the cells (i.e., Econstruct > 10 kPa), it is not neces-
sary to allow a relaxation period (21,22).
13. Safety precautions must be followed when using radiolabeled samples. Gloves
must be worn at all times and monitoring should be performed. A GM tube
counter will detect 35SO4 contamination, but [3H]thymidine contamination can
only be detected using a swab test. This involves wiping the area to be tested with
a damp swab, which is placed in a scintillation vial containing scintillation cock-
tail and counted using a scintillation counter. Count per minute values are com-
pared with background levels, determined from an identical swab, added directly
to the vial.
14. Labeling periods are dependent on the nature of the study. However, certain
points should be considered when conducting radiolabeling experiments with tis-
sue explant or 3D constructs. The radiolabeled molecules must diffuse into the
construct prior to incorporation by cells. Accordingly, very short labeling peri-
322 Lee and Knight
ods are not recommended as the radiolabeled molecules may not have equili-
brated in the construct. Previous studies have demonstrated that [3H]thymidine
takes between 4 and 8 h to equilibrate in 3% agarose constructs of 5-mm diam-
eter and 5-mm height (27). It should be noted that an extended labeling period or
the use of labeled medium with a high radioactivity may lead to DNA or cellular
damage and cause anomalous results. For example, the incorporation of high con-
centrations of [3H]thymidine into DNA during S-phase may lead to DNA dam-
age, which will activate cellular repair mechanisms involving further DNA
synthesis not related to cell proliferation. Further DNA damage is induced, and a
cascade of nonproliferative DNA synthesis may occur, resulting in abnormally
high [3H]thymidine incorporation values.
15. After extended culture periods, digestion at 37°C may not be sufficient to break
up collagenous matrix. In this case, a further incubation period of 1 h at 60°C is
recommended. Note that incubation at 60°C will inactivate the agarase, so the
37°C incubation must be included first to digest the agarose.
16. Please note that Hoechst 33258 binds to DNA and may act as a mutagen. Gloves
must be worn at all times.
References
1.
1 Sah, R., Kim, Y. J., Doong, J. Y. H., Grodzinsky, A. J., Plaas, A. H. K., and
Sandy, J. D. (1989) Biosynthetic response of cartilage explants to dynamic com-
pression. J. Orthop. Res. 7, 619–636.
2.
2 Kim, Y. J., Sah, R. L. Y., Grodzinsky, A. J., Plaas, A. H. K., and Sandy, J. D.
(1994) Mechanical regulation of cartilage biosynthetic behaviour: physical
stimuli. Arch. Biochem. Biophys. 311, 1–12.
3.
3 Vunjak-Novakovic, G., Martin, I., Obradovic, B., et al. (1999) Bioreactor culti-
vation conditions modulate the composition and mechanical properties of tissue
engineered cartilage. J. Orthop. Res. 17, 130–138.
4.
4 Mauck,R. L., Soltz, M. A., Wang, C. C., et al. (2000) Functional tissue engineer-
ing of articular cartilage through dynamic loading of chondrocyte-seeded agarose
gels. J. Biomech. Eng. 122, 252–260.
5. Guilak, F., Butler, D. L., and Goldstein, S. A. (2001) Functional tissue engi-
neering: the role of biomechanics in articular cartilage repair. Clin. Orthop. 391,
295–305.
6.
6 Urban, J. P .G. (1994) The chondrocyte: A cell under pressure. Br. J. Rheumatol
33, 901–908.
7.
7 Heath, C. A. and Magari, S. R. (1996) Mechanical factors affecting cartilage
regeneration in vitro. Biotechnol. Bioeng. 50, 430–437.
8. Guilak, F., Sah, R., and Setton, L.A. (1997) Basic Orthopaedic Biomechanics
(Mow, V. C. and Hayes, W. C., eds.), Lippincott-Raven, Philadelphia, PA, pp.
179–207.
9.
9 Benya, P. D. and Shaffer, J. D. (1982) Dedifferentiated chondrocytes reexpress
the differentiated collagen phentotype when cultured in agarose gels. Cell 30,
215–224.
10. Aydelotte, M. B., Schumacher, B. L., and Kuettner, K. E. (1990) Methods in Car-
tilage Research. (Maroudas, A. and Kuettner, K. E., eds.), Academic, London,
UK, pp. 90–92.
11.
11 Hauselmann, H. J., Fernandes, R. J., Mok, S., et al. (1994) Phenotypic stability of
bovine articular chondrocytes after long-term culture in alginate beads. J. Cell
Sci. 107, 17–27.
12.
12 Knight, M. M., Ghori, S. A., Lee, D. A., and Bader, D. L. (1998) Measurement of
the deformation of isolated chondrocytes in agarose subjected to cyclic compres-
sion. Med. Eng. Phys. 20, 684–688.
13.
13 Lee, D. A., Knight, M. M., Bolton, J. F., Idowu, B. D., Kayser, M. V., and Bader,
D. L. (2000) Chondrocyte deformation within compressed agarose constructs at
the cellular and sub-cellular levels. J. Biomech. 33, 81–95.
14.
14 Lee, D. A., Noguchi, T., Knight, M. M., O’Donnell, L. B., and Bader, D. L. (1997)
Differential metabolic response of superficial and deep zone chondrocytes to com-
pressive strain, in Transactions of the 42nd Annual Meeting of the Orthopaedic
Research Society 22, abstract.
15.
15 Lee, D. A., Noguchi, T., Knight, M. M., O’Donnell, L., Bentley, G., and Bader, D.
L. (1998) Response of chondrocyte subpopulations cultured within unloaded and
loaded agarose. J. Orthop. Res. 16, 726–733.
16. Lee, D. A., Noguchi, T., Frean, S. P., Lees, P., and Bader, D. L. (2000) The influ-
ence of mechanical loading on isolated chondrocytes seeded in agarose constructs.
Biorheology 37, 149–161.
17.
17 Buschmann, M. D., Gluzband, Y. A., Grodzinsky, A. J., and Hunziker, E. B.
(1995) Mechanical compression modulates matrix biosynthesis in chondrocyte/
agarose culture. J. Cell Sci. 108, 1497–1508.
18.
18 Guilak, F., Ratcliffe, A., and Mow, V. C. (1995) Chondrocyte Deformation and
local tissue strain in articular cartilage: A confocal microscopy study. J. Orthop.
Res. 13, 410–421.
19.
19 Knight, M. M., Lee, D. A., and Bader, D. L. (1998) The influence of elaborated
pericellular matrix on the deformation of isolated articular chondrocytes cultured
in agarose. Biochim. Biophys. Acta 1405, 67–77.
20.
20 Knight, M. M., Ross, J. M., Sherwin, A. F., Lee, D. A., Bader, D. L., and Poole, C. A.
(2001) Chondrocyte deformation within mechanically and enzymatically extracted
chondrons compressed in agarose. Biochim. Biophys. Acta 1526, 141–146.
21.
21 Bader, D. L., Ohashi, T., Knight, M. M., Lee, D. A., and Sato, M. (2002) Deforma-
tion properties of articular chondrocytes: a critique of three separate techniques.
Biorheology 39, 69–78.
22.
22 Knight, M. M., van de Breevaart Bravenboer, J., Lee, D. A., van Osch, G. J. V.
M., Weinans, H., and Bader, D. L. (2002) Cell and nucleus deformation in com-
pressed chondrocyte-alginate constructs: temporal changes and calculation of cell
modulus. Biochim. Biophys. Acta 1570, 1–8.
23.
23 Roberts, S. R., Knight, M. M., Lee, D. A., and Bader, D. L. (2001) Mechanical
compression influences intracellular Ca2+ signaling in chondrocytes seeded in
agarose constructs. J. Appl. Physiol 90, 1385–1391.
324 Lee and Knight
24. Knight, M. M., Lee, D. A., and Bader, D. L. (1996) Distribution of chondrocyte
deformation in compressed agarose gel using confocal microscopy. J. Cell. Eng.
1, 97–102.
25.
25 Bachrach, N. M., Valhmu W. B., Stazzone, E., Ratcliffe, A., Lai, W. M., and
Mow, V.C. (1995) Changes in proteoglycan synthesis of chondrocytes in articu-
lar cartilage are associated with the time-dependent changes in their mechanical
environment. J. Biomech. 28, 1561–1569.
26. Knight, M. M. (1997) Deformation of isolated articular chondrocytes cultured in
agarose constructs. PhD Thesis, University of London, London, UK.
27. Lee, D. A. and Bader, D. L. (1997) Compressive strains at physiological frequen-
cies influence the metabolism of chondrocytes seeded in agarose. J. Orthop. Res.
15, 181–188.
Physical Stimulation of Cartilage In Vitro 325
22
In Vitro Physical Stimulation
of Tissue-Engineered and Native Cartilage
Kelvin W. Li, Travis J. Klein, Kanika Chawla,

Gayle E. Nugent, Won C. Bae, and Robert L. Sah
Summary
Because of the limited availability of donor cartilage for resurfacing defects in articular sur-
faces, there is tremendous interest in the in vitro bioengineering of cartilage replacements for
clinical applications. However, attaining mechanical properties in engineered cartilaginous con-
structs that approach those of native cartilage has not been previously achieved when constructs
are cultured under free-swelling conditions. One approach toward stimulating the development
of constructs that are mechanically more robust is to expose them to physical environments that
are similar, in certain ways, to those encountered by native cartilage. This is a strategy motivated
by observations in numerous short-term experiments that certain mechanical signals are potent
stimulators of cartilage metabolism. On the other hand, excess mechanical loading can have a
deleterious effect on cartilage. Culture conditions that include a physical stimulation component
are made possible by the use of specialized bioreactors. This chapter addresses some of the
issues involved in using bioreactors as integral components of cartilage tissue engineering and
in studying the physical regulation of cartilage. We first consider the generation of cartilaginous
constructs in vitro. Next we describe the rationale and design of bioreactors that can impart
either mechanical deformation or fluid-induced mechanical signals.
Key Words: Bioreactors; mechanical stress; tissue engineering; cartilage growth; extracel-
lular matrix; compression; perfusion; shear stress.
1. Introduction
One approach to the prevention of end-stage osteoarthritis is to repair articu-
lar defects, especially those that are focal in nature, using cartilaginous tissue
replacements that are engineered in vitro. Emerging tissue-engineering thera-
pies attempt to reconstitute cartilage structure and function by various processes
that typically begin with the in vitro culture of a combination of cells, support-
ing scaffolding materials, and bioactive molecules (1–5). To date, many of these
325
326 Li et al.
cell-based tissue-engineering methods have yielded promising but inconsistent

results after clinical (6,7) or experimental (8–13) implantation.
A variety of methods for accelerating the development of biochemical and
mechanical properties of growing cartilaginous constructs are under current
investigation. One approach has been to subject such constructs to a mechani-
cal environment that is similar to that encountered by native cartilage in vivo.
This is accomplished by reproducing select aspects of physiological loading in
culture systems collectively known as loading chambers or bioreactors. This
strategy is motivated by in vivo and in vitro observations that mechanical sig-
nals can be potent regulators of cartilage metabolism.
To the extent that chondrocytes function as the biological “architects” of car-
tilage in vivo (14), joint loading provides a partial set of instructions. For
example, weight-bearing regions of the joint tend to have the highest proteo-
glycan concentration (15), and the contrast between the articular cartilage in
immobilized and weight-bearing limbs of growing dogs suggests that joint load-
ing plays a stimulatory role in local cartilage thickening (16). Immobilized joints
in experimental animals also exhibit a reversible decrease in proteoglycan
synthesis and content (17–20), whereas moderate exercise stimulates proteo-
glycan synthesis and accumulation (18,21). Thus, chondrocytes monitor their
mechanical environment in vivo and modify the structure and composition of
their extracellular matrix over time to meet perceived functional demands (22).
The precise consequences of joint loading can be studied from various per-
spectives, ranging from the whole joint down to the tissue and molecular length
scales (Fig. 1). Under physiological joint use, articular cartilage is subjected to
a wide range of mechanical loads (22). At the tissue level, the complex
mechanical stresses associated with joint loading and articulation can be
decomposed into compressive and shear stresses, which are imparted on the
successive layers of cartilage (Fig. 1B). From the perspective of individual
chondrocytes, such loading results in coupled electrical, mechanical, and
chemical phenomena that vary in space and time. These phenomena include
solid deformation, changes in pH, streaming potentials, fluid-induced shear,
hydrostatic pressure, and mass transport of ions and soluble factors (23–25)
(Fig. 1C,D). All these factors represent distinct physical signals that may be
important regulators of chondrocyte behavior. Indeed, the role of the matrix in
determining the nature of the chondrocyte response has been highlighted in a
recent review: “Over the past few years it has been recognized that
chondrocytes in cartilage do not respond to load as such but rather to load-
induced changes in the matrix” (26). However, because these potential regula-
tory cues are inherently coupled within intact cartilage matrix, there are
significant challenges in identifying which of the biophysical signals play pri-
mary roles in triggering specific biological responses by chondrocytes.
Fig. 1. Mechanical loading of articular cartilage: length scales. (A) Normal joint
loading and articulation impart (B) compression (σ) and shear (τ) to the articular car-
tilage at joint surfaces. (C) The material properties of the extracellular matrix dictate
the manner by which loading of cartilage generates discrete biophysical signals to
which individual chondrocytes are exposed. (D) Embedded chondrocytes may initiate
intracellular molecular responses to the induced biophysical signals, leading to cell-
mediated alterations in the molecular composition of the matrix. Micrographs and
schematics adapted from refs. 115–118, with permission.
Multiple studies have used in vivo models based on geometrically defined

cartilage explants to elucidate the effects of defined mechanical stimuli on
cartilage biology. Such explant-based approaches have allowed investigators
to confirm and scrutinize in vivo observations that mechanical loading can
affect the cellular processes that contribute to matrix remodeling (see refs. 27
and 28 for review). In general, static compression of cartilage explants, in
which a fixed compression level is applied and maintained over time, typi-
cally leads to dose-dependent inhibition of matrix synthesis (e.g., refs. 29–
33). On the other hand, the application of dynamic compression, which may
correspond to the time-varying stresses generated by normal joint activity,
328 Li et al.
can stimulate biosynthesis when certain threshold levels of frequency and

amplitude are exceeded (33–36). Under such loading conditions, the stimu-
lated cells appear to be localized to regions predicted to have high fluid flow
(37,38). These responses to mechanical compression appear to be reproduced
when chondrocytes are isolated from their matrix, cultured in three-dimen-
sional (3D) gels, and then subjected to similar types of compression (39–41),
but sometimes only after a significant amount of matrix has accumulated
(39,41). Isolated chondrocytes also retain the ability to respond metabolically
to different types of mechanical forces, a finding supported by a growing body
of research in which they are subjected to distinct components of the in situ
mechanical environment. In chondrocytes cultured in monolayer, glycosami-
noglycan (GAG) biosynthesis was stimulated by cyclic tensile strain (42),
fluid-induced shear (43), and extended high-frequency pressurization (44).
In contrast, when compression is applied to cartilage explants at high strain
magnitude or strain rate, injurious effects can be induced. This type of loading
can mimic cartilage impact. The resultant effects have implications for the ini-
tiation and progression of cartilage damage and deterioration. Matrix damage
in the form of collagen denaturation can occur with a compressive strain mag-
nitude as low as 25% at a relatively low strain rate of 0.1/s in unconfined com-
pression (45), even without chondrocyte death. Cell death, including apoptosis,
varies with a number of mechanical loading parameters, including strain mag-
nitude, rate, and duration or number of cycles, with strain magnitude being the
major effector. Strain magnitudes as low as 30% can result in cell death, if the
strain rate is sufficiently high (46,47). With an increase in applied strain mag-
nitude, chondrocyte viability decreases (46–50), even while the same strain
rates are held constant (46). Once injured, cell death can occur either in the
short- or the long-term, with cell death sometimes increasing with time after
loading (46,51).
Thus, the mechanical signals generated by both normal and injurious joint
loading can regulate various cellular processes within chondrocytes in cartilage.
The same cellular processes may also underlie the generation of de novo extra-
cellular matrix during in vitro tissue engineering procedures, with upregulation
of such processes promoting chondrogenesis. The latter observations form the
rationale for bioreactor approaches that attempt to exploit appropriate mechani-
cal signals for the enhancement of engineered cartilaginous constructs during
extended culture periods (weeks to months). The intent of such bioreactors is to
apply mechanical stimulation to cartilaginous constructs so that the induced cel-
lular activities result in increased accumulation and assembly of matrix compo-
nents and concomitant changes in the functional mechanical properties of the
tissue. These bioreactors are similar in many respects to chambers that have
been used to impart injurious stimuli to cartilage.
Fig. 2. Tissue engineering strategies to repair an articular cartilage defect in the

femoral condyle.
This chapter addresses some of the issues involved in using bioreactors as

integral components of current experimental studies of cartilage tissue engi-
neering and physical regulation of chondrocytes and describes some key steps
in such studies, namely, sample preparation and mechanical stimulation by
solid deformation or fluid-based perfusion. An initial step in cartilage tissue
engineering is the fabrication of cartilaginous constructs prior to incubation
within a bioreactor.
1.1. Sample Preparation
The cartilage bioengineer is faced with several key design choices on the
pathway to fabricating functional cartilaginous constructs (Fig. 2). The first of
these choices is the selection of an appropriate cell population. The cell types
that have been used to generate cartilaginous tissue include chondrocytes (52),
perichondrial (53) and periosteal cells (54), and other pluripotent cells, includ-
ing those derived from bone marrow (55,56) and fat (57). Once the cell type
has been selected, a decision must be made about how to manipulate these cells
prior to construct formation. This is a cell-type specific decision, and manipu-
lations can range from direct use of primary cells (as is possible with
chondrocytes), to expanding the cell number in monolayer or a 3D gel with or
without treatment of growth factors to induce a chondrogenic phenotype (see
Note 1) (58).
After generating a sufficient number of desirable chondrogenic cells, the
next step is to decide what type of scaffold, if any, will be used in which to seed
the cells and form the initial shape of the construct. A popular class of scaf-
folds has been biodegradable polymers such as polyglycolic acid (PGA) (52),
330 Li et al.
polylactic acid (PLA) (59), and their co-polymers (60). These materials pro-
vide initial mechanical support for the cells, and eventually degrade to leave
de novo tissue. Other scaffolds use biopolymers such as collagen (11,61,62)
and hyaluronic acid (63), which are components of the native cartilage matrix.
A third class of scaffold materials is hydrogels (64), which have the possible
advantage of injection into a defect site. Among the commonly used gels are
those polymerized by temperature (agarose) (65), ions (alginate) (66), cross-
linking enzymes (fibrin) (67), and light (poly[ethelyne oxide]) and poly[eth-
ylene glycol]) (68,69). Cartilaginous tissue can be formed without scaffolds
under certain conditions. These include direct seeding of culture-expanded
chondrocytes into a covered defect space (6), micromass culture of chondro-
genic cells (70), high-density culture of chondrocytes from immature animals
(71), and high-density culture of chondrocytes from adults after preculture in,
and recovery from, alginate (5).
When designing construct geometry, it is important to consider bioreactor
stimulation, laboratory analysis after in vitro culture, and future clinical appli-
cation of the engineered cartilage. Having a well-defined geometry, such as a
cylindrical disk, can ease the design of bioreactors and allow for simplified
models of the solid- and fluid-mechanical environment (72). Additionally, con-
trolling the shape of the developing constructs can allow for more straightfor-
ward biochemical and mechanical analysis of the newly formed tissue
following culture, without the need for excessive sample manipulation.
Although these geometries facilitate tissue formation and subsequent analysis,
the application of engineered tissue to cartilage defects may require more com-
plex geometries. If anatomically shaped implants are desired, it is possible to
combine 3D imaging techniques such as magnetic resonance imaging (MRI)
and sophisticated manufacturing procedures such as 3D printing (73) and pre-
cision molding (74) to generate replacements for very specific defects and pos-
sibly even whole joints (75).
Another important choice that significantly affects the growth of the engi-
neered cartilaginous construct is the method of cell seeding. With a preformed
scaffold, uniformity and efficiency of seeding are key design goals in the suc-
cess of tissue generation (76). Static (passive) seeding, in which cells in a dense
suspension are allowed to sediment upon (and to a certain extent into) the scaf-
fold, is typically inefficient and results in a high cell density on the periphery
of the constructs but a low cell density in the center. Dynamic seeding, either
in a well-plate on an orbital shaker or in a spinner flask or rotating bioreactor,
enhances both the efficiency and the uniformity of cells seeded into the scaf-
fold (77). If a hydrogel is used, vigorous mixing of the cell/hydrogel suspen-
sion, prior to polymerization, facilitates cell dispersion. Also important is
minimizing the time needed for polymerization, especially if the curing pro-
cess has cytotoxic effects. When an exogenous scaffold is omitted (e.g.,

micromass [70] or high-density culture [5,71]), the main decisions are the cell
seeding density and the amount of time allowed for the cells to form a cohesive
mass. With such methods, there is no scaffold to provide volume filling so a
very high area density of cells (106–107 cells/cm2) is typically formed after cell
settling, which is generally an overnight process.
1.2. Mechanical Stimulation In Vitro
The acquisition of mechanical properties in engineered cartilage that approach
those of native cartilage does not appear to occur when the tissue is cultured
under free-swelling conditions, even for long durations (78,79). After 12 mo,
cartilaginous constructs in free-swelling culture attained compressive moduli that
were only 10% of those of the native tissue from which the seeded chondrocytes
had been derived (79). Although such free-swelling culture conditions provides
the investigator with some control over the application of chemical signals (e.g.,
via addition of growth factors and serum, levels of carbon dioxide, and so on),
mechanically proactive culture conditions offer more precise control of the envi-
ronmental signals, both chemical and mechanical, to which the cells are exposed.
Culture conditions that include a mechanical component are made possible by
the use of specialized loading chambers, such as bioreactors, in which these
aspects of the in vitro culture environment can be regulated.
A number of design criteria are common to all culture approaches that rely on
the application of mechanical stimuli. Above all, the loading chambers used must
maintain an appropriate environment for tissue culture; thus they must satisfy
classical tissue culture criteria (80), including control of ambient temperature
and humidity, sufficient exchange of nutrients, gasses, and regulatory factors,
sterility, and biocompatibility. Furthermore, bioreactors must be designed to
allow the reproducible application of physical stimuli. In this section, we con-
sider details regarding specific modalities of mechanical loading that can be
accomplished with the implementation of a bioreactor culture strategy for the
tissue engineering of cartilage.
1.2.1. Solid Deformation
During normal ambulation, articular cartilage is subjected to a complex pat-
tern of joint loading that can be decomposed, in general, into time-averaged
(static) and time-varying (dynamic) components. Studies indicating that
dynamic compression, at certain amplitudes and frequencies, is capable of
stimulating the metabolic activities of chondrocytes in cartilage explants in
short-term in vitro experiments (28) corroborate long-time observations in vivo
(22) and suggest one modality by which engineered cartilage can be subjected
in the long term to the stimulatory influence of physiologic loading (Fig. 3A).
332 Li et al.
Fig. 3. Various loading configurations for mechanical deformation of cartilaginous

constructs. (A) Both unconfined and confined compression can be applied using the
same incubator-housed actuator apparatus, depending on the sample fixture and platen
configuration. (B) Tissue shear deformation can be imposed by positioning cartilagi-
nous constructs in contact with culture plate surfaces that rotate relative to each other.
Note that for emphasis, the direction of applied shear is shown to be lateral in the
figure; in actuality, shear deformation occurs in the direction tangential to the rotation
path, which would be into and out of the plane of the page.
In general, accumulation of extracellular matrix de novo involves a balance

between a number of biological processes, including matrix synthesis, post-trans-
lational processing, and extracellular assembly, deposition, and degradation.
Although the stimulatory effects of moderate dynamic compression on over-
all matrix synthesis have been shown, such loading may reduce the efficiency
at which newly synthesized matrix molecules are deposited in cartilage
explants, with the induced fluid convection causing an increase in the release of
such molecules into medium (81,82), partially counteracting the increased syn-
thesis. Therefore, for long-term accumulation and assembly of cartilaginous
matrix, it may be beneficial to introduce periods of load-free, free-swelling cul-
ture between controlled duty cycles of dynamic loading, especially since the
release of compression exerts stimulatory effects on protein synthesis that can
persist for hours (33,83). Bioreactor approaches that include both duty cycles
of dynamic compression and periods of load-free culture have been introduced

recently (65,84,85). In one such study (65), chondrocytes from 3–5 mo-old
bovines were seeded into agarose hydrogels. Once introduced into the
bioreactor chamber, cartilaginous constructs underwent 10% dynamic compres-
sion at a frequency of 1 Hz for a period of 3 h per day. At the end of 21 d of
culture, the loaded constructs exhibited an equilibrium modulus that was approx
six times greater than that of their unloaded counterparts and approached 25%
of that of native calf cartilage (86), demonstrating the potential value of dy-
namic loading in the engineering of functional cartilage replacements.
Other modalities of cartilage loading have been investigated as possible regu-
lators of matrix synthesis by the chondrocytes in cartilage explants. Although
dynamic compressive loading of a relatively low amplitude has been associated
with cartilage stimulation and maintenance (28), excessive loading to the point
of impact has been implicated in initiating a cascade of injurious biological
processes, such as chondrocyte necrosis and apoptosis and loss of matrix mol-
ecules, during subsequent culture of cartilage explants (46,87,88). On the other
hand, the application of dynamic shear deformation to cartilage tissue (Fig. 3B)
stimulated both proteoglycan and protein synthesis (89). The effects of such
macroscopic shear deformation did not vary with shear amplitude or frequency.
This mode of mechanical loading, in particular, provides novel mechanistic
information about the signals to which chondrocytes respond, since under
dynamic shear deformation, cartilage undergoes dynamic deformation in the
absence of significant induction of fluid flow. Such insights into specific
mechanical regulators of cartilage have significant tissue engineering implica-
tions: although the role of appropriate mechanical loading in maintaining
healthy cartilage is well documented, there are many mechanical strategies for
enhancing matrix accumulation in a tissue-engineered construct.
A variety of specialized loading chambers for applying various forms of
mechanical loading have been developed for short-term studies using cartilage
explants (e.g., ref. 90). The design of incubator-housed bioreactors for the long-
term application of mechanical deformation requires some modifications to
accommodate both a highly humidified atmosphere and the extended culture
periods that necessitate strict standards for sterility and biocompatibility. In
general, apparati designed for applying mechanical loads in extended culture
must include a culture chamber component that satisfies traditional criteria for
tissue culture environments and an actuator component to impart mechanical
deformation to samples in a controlled and reproducible manner. The appara-
tus can be designed to impose either a prescribed displacement or a prescribed
load. It may be desirable either to measure or to predict theoretically the other
parameter, since prior knowledge of the range in these parameters is critical in
determining whether the apparatus has enough stiffness to resist deflection.
334 Li et al.
It is also important to consider that the nature of physical stimulation may

require adjustment over time in culture, as the growing tissue-engineered con-
struct will probably undergo concomitant changes both in geometry and in
material properties. These changes in the developing tissue can potentially
alter the nature of the physical signals that are transmitted to indwelling cells
by the bioreactor and must be considered in any long-term loading protocol
design. For instance, a given applied load may be associated with progres-
sively less strain as constructs under dynamic compression acquire increased
mechanical stiffness. Also, in the case of tissue shear deformation, slippage
between the platen and the tissue sample can also occur at higher strain ampli-
tudes. For these reasons, it may be desirable to monitor aspects of the applied
mechanical signals. Such instrumentation can be incorporated into the
bioreactor design as part of a feedback control system.
1.2.2. Fluid-Based Perfusion
Perfusion bioreactors, including rotating-wall and direct perfusion
bioreactors, represent common forms of bioreactors currently used in cartilage
tissue engineering. One of the major design goals of such bioreactors is to
promote exchange of nutrients, oxygen, and metabolites between the tissue
and culture medium. In addition, the concomitant application of fluid-induced
shear stress, compression, hydrostatic pressure, and other factors may provide
necessary mechanical signals for enhancing cartilaginous tissue formation. In
general, perfusion bioreactors are connected to a medium reservoir that pro-
vides a continuous supply of fresh or recycled medium at a controlled flow rate
to cartilaginous constructs.
Rotating-wall bioreactors provide a simulated microgravity environment for
suspension culture of constructs, while providing laminar flow to enhance mass
transfer and to provide mechanical signals (Fig. 4A). Although most of the
fluid flows around the constructs, and the flow through the constructs is com-
plex, these vessels have been quite successful in stimulating cartilaginous
growth and development. Tissue-engineered constructs cultured in rotating-
wall bioreactors, for both long (7-mo) and short (6-wk) periods, show increased
cell proliferation, accumulation of extracellular matrix, particularly GAG and
collagen type II (91–93), and ability to bear compressive load, compared with
constructs cultured under free-swelling conditions.
Direct perfusion bioreactors maintain constructs in the path of the fluid
flow (Fig. 4B) and therefore have the advantage of a more easily defined
flow through the constructs. These bioreactors subject constructs to signifi-
cant physical forces during perfusion culture, including quasi-static compres-
sion (94) and fluid shear stress, which can directly influence the properties of
tissue-engineered constructs (95). Direct perfusion affects matrix metabo-
Fig. 4. Culture configurations for perfusion of cartilaginous constructs. (A) In

rotating wall bioreactors, growing constructs become suspended in the path of the
fluid flow induced by the relative motion of the cylindrical walls. The magnified image
is a schematic depiction of the fluid path of least resistance, in which fluid flows around
the cultured constructs. (B) Constructs can also be positioned within perfusion cham-
bers, as shown with accompanying peristaltic pump, media reservoirs, and connecting
tubing. In these chambers, the constructs obstruct the path of the fluid flow, and fluid
is driven through the cartilaginous constructs.
lism, construct growth, and resultant mechanical properties of constructs.

Both intermittent (96) and continuous fluid flow (2,82,95,97,98) applied at
linear velocities of 0.1–100 µm/s to tissue engineered constructs cultured for
up to 4 weeks leads to increases in DNA, GAG, and hydroxyproline content
compared with static controls (2,82,97). The timing of perfusion stimuli may
be important, as application early in culture can inhibit GAG synthesis and
accumulation, whereas perfusion later in culture can exhibit the opposite
effect (82).
The following sections describe first the generation of cartilaginous con-
structs in vitro and then the setup of bioreactors that can impart either mechani-
cal deformation or fluid-induced mechanical signals. The intent of some
bioreactors is to apply mechanical stimulation over such time that induced cel-
lular activities result in increased accumulation of matrix components and con-
comitant changes in the functional mechanical properties of cartilaginous
constructs. The intent of others is to induce graded levels of cartilage injury.
336 Li et al.
2. Materials
2.1.1. Cartilage Harvest and Explant Preparation
1. Bovine knee.
2. Table vise.
3. Sterile instruments: scalpel, forceps, squirt bottle, saw blade.
4. Autopsy saw.
5. Sterile phosphate-buffered saline (PBS) at pH 7.4.
6. Penicillin/streptomycin/Fungizone (P/S/F) 100X stock solution, containing
10,000 U penicillin, 10,000 µg streptomycin, and 25 µg amphotericin B per mL
(Invitrogen).
7. Sledge microtome (Microm, cat. no. HM440E).
8. Cylindrical punch.
2.1.2. Chondrocyte Isolation

1. Culture medium (DMEM/F12+).
a. Dulbecco’s modified Eagle’s medium/Ham’s F12 (DMEM/F12).
b. 0.4 mM L-proline from 100 mM solution, made from powder (Sigma-Aldrich)
and filter-sterilized.
c. 2 mM L-glutamine from 100X stock (Invitrogen).
d. 0.1 mM nonessential amino acids from 100X stock (Sigma-Aldrich).
e. P/S/F, diluted 1:100.
2. Fetal bovine serum (FBS; HyClone).
3. Protease type XIV (Sigma-Aldrich).
4. Collagenase P (Roche).
5. Sterile sample cup.
6. Sterile stir bar.
7. 0.22-µm Sterile syringe filter.
8. Cell strainers (40 µm and 70 µm).
9. Trypan blue.
10. Hemacytometer (Neubauer).
11. Centrifuge capable of 750g.
2.1.3. Scaffold Seeding

1. Chondrogenic cells.
2. PLA scaffold (Kensey Nash, DRILAC Cube).
3. Dermal punch.
4. Orbital shaker.
2.1.4. Free-Swelling Construct Culture

1. Ascorbic acid, 25 µg/mL.
2. Multiwell tissue culture plates.
3. DMEM/F12+.

1. Custom-designed compression or shear bioreactor constructed from
biocompatible materials, such as polysulfone or polycarbonate (e.g., Fig. 3; see
Note 2 for some design considerations).
2. Mechanical actuator capable of imparting desired load or displacement (e.g.,
linear stepper motors, cam-driven motors, or mechanical testing machines; see
Note 3).
3. Standard incubator for housing bioreactor apparatus to create a culture-like envi-
ronment (5% CO2, 95% water-saturated air at 37°C; see Note 4).
4. Cartilaginous constructs or cartilage explants of defined geometries (e.g.,
2–5 mm thick).
5. Medium components.
a. DMEM/F12+.
b. FBS.
c. Ascorbic acid.

1. Custom-designed perfusion bioreactor constructed from biocompatible materi-
als, such as polysulfone or polycarbonate (e.g., Fig. 4B; see Note 5 for some
design considerations).
2. Multichannel peristaltic pump (Cole-Parmer Masterflex 9770-50) (see Note 6).
3. Media reservoir that allows gas exchange.
4. Gas-permeable tubing, such as silicone.
5. Individual polyglycolic acid scaffolds (PGA mesh with 97% porosity and 45-mg/
cm3 density; Smith & Nephew) of known diameter and thickness to be seeded
with cells and placed in individual perfusion bioreactors.
6. Chondrogenic cells in suspension.
7. Medium components.
a. DMEM/F12+.
b. FBS.
c. Ascorbic acid.
3. Methods
The following is an example of steps that may be used to fabricate and
mechanically stimulate cartilaginous tissue. First, a method is provided for
obtaining geometrically defined cartilage explants that can serve either as con-
trol samples for tissue engineering studies, or as samples with which to study
the effects of various forms of in vitro stimulation. Then a method to produce
cartilaginous constructs, consisting of PLA disks seeded with bovine articular
chondrocytes, is provided.
338 Li et al.
3.1.1. Cartilage Harvest and Explant Preparation (99)

1. Obtain an intact bovine knee still surrounded with beef from a local abattoir.
2. Mount the femur in a table vise. Open knee surgically. Use sterile instruments to
remove soft tissue and expose the cartilage surfaces of the distal femur. Keep
cartilage surfaces irrigated with PBS supplemented with P/S/F.
If cartilage explants are to be formed, steps 3–5 are useful. Otherwise, carti-
lage can be cut off with a scalpel and minced into approx 1-mm3 pieces.
3. Cut osteochondral blocks from the patellofemoral groove and/or femoral
condyles using an autopsy saw under copious irrigation with PBS+P/S/F (see
Note 7). Blocks should be rectangular and sized to allow for a relatively flat
cartilage surface (~1 × 1 cm2) and should contain approx 1 cm of bone. Store
blocks in a sterile sample cup with PBS+P/S/F at 4°C.
4. Mount the blocks on a sledge microtome by gripping the bone such that the car-
tilage surface is horizontal. Keep blade and cartilage hydrated with PBS+P/S/F.
a. Incrementally (50–100 µm steps) raise the block until the blade just touches
the cartilage surface.
b. Move the block to the desired thickness (0.2–1.0 mm) and cut the cartilage to
obtain slices. Successive cartilage slices can be taken down to the bone (see
Note 8).
5. Trim the slices to fit culture chamber or testing device (see Note 9).
3.1.2. Chondrocyte Isolation (100)

1. Obtain cartilage slices as above, or dissect cartilage off joint directly, mince into
pieces, and store them in a sterile sample cup with PBS+P/S/F at 4°C.
2. Digest cartilage enzymatically.
a. Make medium for enzymatic digestion and cultures.
i. To DMEM/F12, add sterile L-proline to 0.4 mM, L-glutamine to 2 mM,
nonessential amino acids to 0.1 mM, and P/S/F (1:100) to final concentra-
tions of 100 U/mL, 100 µg/mL, and 0.25 µg/mL, respectively. This
medium is referred to hereafter as DMEM/F12+.
ii. Add FBS to DMEM/F12+ (5% final concentration). Make 100 mL for
digestion of 3g or less of cartilage.
iii. Dissolve protease type XIV (0.4%) and collagenase P (0.05%) separately
in DMEM/F12+ with 5% FBS. Prepare 6 mL of each enzyme solution per
gram of cartilage. Use a 0.22-µm syringe filter to sterilize the solutions
and keep warm in a 37°C incubator.
b. Add cartilage to the sample cup with protease solution and incubate for 1 h
while stirring using a magnetic stir plate (see Note 10). Remove medium, and
rinse with 50 mL DMEM/F12+ with 5% FBS.
c. Add collagenase medium to the sample cup with the cartilage slices and incu-
bate for 16 h while stirring (see Note 11).
d. Transfer the solution to a sterile 50-mL centrifuge tube through a 70-µm cell
strainer. Rinse an additional 25 mL of 4°C PBS through the strainer.
e. Centrifuge at 750g for 5 min. Remove and discard the supernatant. Repeat
this step an additional two times, each time resuspending the pellet before
centrifugation. This removes collagenase activity.
f. Resuspend the cell pellet in DMEM/F12+ with 10% FBS, and filter through a
40-µm cell strainer to obtain a single cell suspension.
g. Measure the volume of the cell suspension, and save 1 drop in a well-plate.
Mix 28 µL of the cell suspension with 7 µL of trypan blue. Pipet 10 µL into
each side of a hemocytometer. Count the viable (clear) and dead (blue) cells
in the main grid of each side.
h. Multiply the average number of live cells by 12,500 to obtain a cell density
(cells/mL). Multiply the cell density by the suspension volume to obtain the
total number of primary chondrocytes isolated (see Note 1).
3.1.3. Scaffold Seeding (101)

1. Concentrate the cells to 25 million cells/mL by pelleting the cells and aspirating
off the appropriate volume of medium.
2. Create sterile cores of nonwoven PLA mesh using a 3.7-mm-diameter dermal
punch on prefabricated polymer cubes. Cut these cores into 3-mm-long segments.
3. Prewet the scaffolds in DMEM/F12+ with 10% FBS for 1 h in a 6-well plate,
while revolving at 100 rpm on an orbital shaker.
4. Place PLA scaffolds into individual 50-mL tubes. Seed 80 µL of the cell sus-
pension in each tube. Place the tubes on ice and on an orbital shaker. Shake the
tubes at 100 rpm for 1 h to seed the scaffolds efficiently and form cell-laden
constructs.
3.1.4. Free-Swelling Construct Culture

1. Place the scaffolds in individual wells of a non-tissue culture-treated 24-well
plate. Add 4 mL of medium (DMEM/F12+ with 10% FBS and 25 µg/mL ascor-
bic acid) to each well.
2. Incubate the plate at 37°C in a humidified incubator for 2 d to allow cells to
attach firmly to the scaffold.
3. Collect and freeze a portion of the medium for future analysis.
4. Culture for a given period of time (e.g., 2 wk), with medium changes every other
day. Collect and freeze medium for future analysis (see Note 12).

Once a bioreactor has been designed for the application of solid deforma-
tion, the culture protocol can proceed in a relatively straightforward manner.
What follows are the general steps to synthesize cartilaginous constructs under
the influence of mechanical deformation (e.g., refs. 33 and 65; Fig. 3):
1. Prepare cartilaginous constructs or cartilage explants of defined thickness (2–5 mm).
2. Determine the specifics of the loading protocol (see Note 13).
340 Li et al.
a. Configuration (e.g., confined or unconfined compression, or shear loading).

b. Amplitude (physiological or injurious levels).
c. Frequency.
d. Duty cycles and/or periods of load-free culture.
3. Connect bioreactors to mechanical actuators.
4. House the apparatus in a standard incubator (5% CO2, 37°C).
5. Maintain samples in bioreactors containing 2 mL per million cells of medium,
such as DMEM/F12+ supplemented with 10% FBS and 50 µg/mL ascorbate.
6. Maintain free-swelling controls in sterile containers in parallel culture.
7. Change medium three to seven times a week, at regular intervals.

Varying levels of perfusion can be used to modulate matrix content and can
also be used to enhance growth of tissue-engineered cartilage in vitro. Below is
an example of one design approach for generating tissue-engineered articular
cartilage under conditions of direct perfusion culture using PGA scaffolds
seeded with bovine articular cartilage chondrocytes (Fig. 4B) (2):
1. Connect bioreactors to media reservoir using gas-permeable tubing such as sili-
cone.
2. Connect bioreactors to a multichannel peristaltic pump with silicone tubing.
3. House apparatus in a standard incubator (5% CO2, 37°C).
4. Perfusion seeding (82).
a. Press-fit PGA scaffolds of known diameter and thickness into individual per-
fusion bioreactors.
b. Inject chondrocytes in suspension in DMEM/F12+ supplemented with 10%
FBS and 25 µg/mL ascorbic acid for a targeted seeding density of 25 million
cells/cm3 scaffold material.
c. Seed the cells at a flow rate of 2.5 mL/min (corresponding to a linear fluid
velocity of ~500 µm/s) for 5 h.
d. Decrease the flow rate after seeding to 0.05 mL/min (~10 µm/s) and leave the
constructs overnight (for 1-d seeding) or longer, making sure to change the
medium every 2–3 d.
6. Once the initial growth and seeding is complete, constructs in bioreactors can be
subjected to continuous or intermittent flow for the remainder of the culture
period (see Note 14).
3.3. Discussion
In clinical practice, the biological resurfacing of regions of damaged articular
cartilage may involve the transplantation of site-matched osteochondral allografts
(102–107). Such a transplantation material may immediately withstand the nor-
mal mechanical demands placed on joint cartilage, a challenging environment
that would disrupt tissue with substantially inferior mechanical properties.
Because of the limited availability of donor tissue, the engineering of mechani-

cally functional cartilage replacements for biological resurfacing has tremen-
dous clinical implications. This chapter has reviewed some of the promising
methodologies that are currently being investigated as aids to this ultimate goal
of generating tissue-engineered cartilaginous tissue that will become, or remain,
functional in the recipient. Some of the same methodologies can be used to study
the response of cartilage explants to injurious compression. In summary, the in
vitro development of cartilage-like material properties in cartilaginous constructs
may require exposure to appropriate mechanical signals that essentially “precon-
dition” the growing tissue in preparation for the mechanical challenges in vivo.
4. Notes
1. Monolayer expansion of cells tends to cause chondrocytes to dedifferentiate into
a more fibroblastic phenotype, but further treatment in a three-dimensional gel
can induce redifferentiation (108,109).
2. Bioreactor design.
a. Materials such as polysulfone or polycarbonate are convenient choices for
bioreactor materials because, in addition to being relatively biocompatible,
they can be fabricated by either injection molding or machining, and also can
be sterilized in an autoclave.
b. Particular care must be taken in selecting biocompatible materials for the plat-
ens that come into contact with samples and transmit mechanical loads. Such
platens can be permeable or impermeable, depending on the specifics of the
loading protocol.
3. Mechanical actuator design.
a. These actuators can be based on cam-driven (65) or linear stepper motors (90)
and may induce deformations directly (displacement control [90]) or via
spring-loaded rods (load control [110,111]).
b. Alternatively, the chambers may be attached to a mechanical testing machine
(33).
c. If load-free cycles are to be inserted between duty cycles, it may be necessary
to disengage the culture chambers from the actuators during periods of free-
swelling culture.
d. It may be desirable to monitor aspects of the applied mechanical signals in
real time. Such instrumentation can be incorporated into the bioreactor design
as part of a feedback control system.
e. The apparatus (bioreactor and mechanical actuator) should be designed to
withstand the humidified environment of a standard incubator (e.g., rust-
proof, with motorized parts properly sealed).
4. An alternative to incubator-housing the bioreactor apparatus is to circulate
warmed, pH-controlled medium through an external apparatus to create a cul-
ture-like environment.
342 Li et al.
5. Bioreactor design (e.g., ref. 112). Perfusion-based bioreactors should be designed

so that there is a tight fit between the edge of the construct and the inner wall of
the bioreactor. This prevents fluid from flowing around the periphery of the con-
struct and directs flow through the construct itself.
6. Perfusion system design.
a. A multichannel pump is useful if multiple bioreactors are to be subjected to
perfusion in parallel flow paths.
b. Peristaltic pumps may require the use of an inline damper to reduce pressure
fluctuations.
c. Flow medium can be recirculated, both for reducing the amount of medium
needed and also to preserve some of the chemical signals, i.e., growth factors
or proteins, secreted by cartilaginous constructs. Fresh medium can also be
mixed with spent medium to serve this purpose.
d. The pump should be capable of generating the hydrostatic pressures that occur
as the prescribed fluid flow rate is forced through constructs that typically
decrease in permeability during culture.
e. Monitoring pressure may be useful.
7. Alternatively, osteochondral cores can be taken with surgical harvesting devices
such as the Osteochondral Autograft Transfer System (Arthrex).
8. The depth of slices, relative to the articular surface, may be important, as there
are zonal variations in the chondrocytes of cartilage (113).
9. A typical configuration is a disk (e.g., 3-mm-diameter × 1-mm-thick), formed by
using a dermal punch (111).
10. Loosely cap the enzyme solutions during the digestion to allow for gas exchange.
11. After the collagenase incubation, there should not be any visible pieces of tissue
remaining, and the medium should be slightly cloudy.
12. Periodically move constructs to a new well-plate to avoid cell outgrowth on the
surface of the plate.
13. Loading protocol.
a. Configuration: although many of the short-term experimental studies on the
effects of compression on cartilage have employed explants loaded in a radi-
ally unconfined configuration (28,33), loading in confined compression
(110,114) has the advantage of rigorously maintaining the radial dimension of
a growing construct. Shear-loaded samples are typically not confined radially.
b. Amplitude: low-amplitude dynamic compression tends to mimic stimulatory
effects of physiologic loading, whereas high strain rate impact loading can
lead to cartilage injury.
c. Frequency: higher loading frequencies (>0.01 Hz [33]) have stimulatory
effects on biosynthesis and may approximate such effects of physiologic load-
ing. However, at high frequency-amplitude combinations, lift-off can occur,
when the samples do not “rebound” quickly enough and when there is no
platen resting on top.
d. Duty cycle: periods of free-swelling culture in between duty cycles of loading
seem to facilitate stabilization of newly synthesized matrix molecules. This
may not be necessary for samples loaded with dynamic tissue shear, since
little fluid convection occurs in that loading configuration (90).
14. Spent medium from seeding or growth perfusion culture may be saved for later
analysis.
References
1.
1 Freed, L. E., Vunjak-Novakovic, G., and Langer, R. (1993) Cultivation of cell-
polymer cartilage implants in bioreactors. J. Cell. Biochem. 51, 257–264.
2. Dunkelman, N. S., Zimber, M. P., LeBaron, R. G., Pavelec, R., Kwan, M., and
Purchio, A. F. (1995) Cartilage production by rabbit articular chondrocytes on
polyglycolic acid scaffolds in a closed bioreactor system. Biotechno.l Bioeng.
46, 299–305.
3. Wakitani, S., Kimura, T., Hirooka, A., et al. (1989) Repair of rabbit articular
surfaces with allograft chondrocytes embedded in collagen gel. J. Bone Joint.
Surg. Br. 71-B, 74–80.
4.
4 Kim, W. S., Vacanti, J. P., Cima, L., et al. (1994) Cartilage engineered in prede-
termined shapes employing cell transplantation on synthetic biodegradable poly-
mers. Plast. Reconstr. Surg. 94, 233–237.
5.
5 Masuda, K., Sah, R. L., Hejna, M. J., and Thonar, E. J.-M. A. (2003) A novel
two-step method for the formation of tissue engineered cartilage: the alginate-
recovered-chondrocyte (ARC) method. J. Orthop. Res. 21, 139–148.
6.
7.
7 Peterson, L., Minas, T., Brittberg, M., Nilsson, A., Sjogren-Jansson, E., and
Lindahl, A. (2000) Two- to 9-year outcome after autologous chondrocyte trans-
plantation of the knee. Clin. Orthop. 374, 212–234.
8. Breinan, H. A., Minas, T., Barone, L., et al. (1998) Histological evaluation of
the course of healing of canine articular cartilage defects treated with cultured
autologous chondrocytes. Tissue Eng. 4, 101–114.
9.
9 Brittberg, M., Nilsson, A., Lindahl, A., Ohlsson, C., and Peterson, L. (1996)
Rabbit articular cartilage defects treated with autologous cultured chondrocytes.
Clin. Orthop. 326, 270–283.
10. Nehrer, S., Breinan, H. H., Ashkar, S., et al. (1998) Characteristics of articular
chondrocytes seeded in collagen matrices in vitro. Tissue Eng. 4, 175–183.
11.
11 Sams, A. E. and Nixon, A. J. (1995) Chondrocyte-laden collagen scaffolds for
resurfacing extensive articular cartilage defects. Osteoarthritis Cartilage 3,
47–59.
12.
12 Shortkroff, S., Barone, L., Hsu, H. P., et al. (1996) Healing of chondral and
osteochondral defects in a canine model: the role of cultured chondrocytes in
regeneration of articular cartilage. Biomaterials 17, 147–154.
13.
13 Wakitani, S., Goto, T., Young, R. G., Mansour, J. M., Goldberg, V. M., and Caplan,
A. I. (1998) Repair of large full-thickness articular cartilage defects with allograft
articular chondrocytes embedded in a collagen gel. Tissue. Eng. 4, 429–444.
344 Li et al.
14.
14 Muir, H. (1995) The chondrocyte, architect of cartilage. Biomechanics, struc-
ture, function and molecular biology of cartilage matrix macromolecules.
Bioessays 17, 1039–1048.
15.
15 Slowman, S. D. and Brandt, K. D. (1986) Composition and glycosaminoglycan
metabolism of articular cartilage from habitually loaded and habitually unloaded
sites. Arthritis Rheum. 29, 88–94.
16.
16 Kiviranta, I., Jurvelin, J., Tammi, M., Saamanen, A.-M., and Helminen, H. J.
(1987) Weight bearing controls glycosaminoglycan concentration and articular
cartilage thickness in the knee joints of young beagle dogs. Arthritis Rheum. 30,
801–809.
17. Behrens, F., Kraft, E. L., and Oegema, T. R. (1989) Biochemical changes in
articular cartilage after joint immobilization by casting or external fixation.
J. Orthop. Res. 7, 335–-343.
18.
18 Caterson, B. and Lowther, D. A. (1978) Changes in the metabolism of the
proteoglycans from sheep articular cartilage in response to mechanical stress.
Biochim. Biophys. Acta 540, 412–422.
19.
19 Jurvelin, J., Kiviranta, I., Saamanen, A.-M., Tammi, M., and Helminen, H. J.
(1989) Partial restoration of immobilization-induced softening of canine articu-
lar cartilage after remobilization of the knee (stifle) joint. J. Orthop. Res. 7,
352–358.
20 Saamanen, A.-M., Tammi, M., Jurvelin, J., Kiviranta, I., and Helminen, H. J. (1990)
20.
Proteoglycan alterations following immobilization and remobilization in the artic-
ular cartilage of young canine knee (stifle) joint. J. Orthop. Res. 8, 863–873.
21.
21 Kiviranta, I., Tammi, M., Jurvelin, J., Saamanen, A. M., and Helminen, H. J. (1988)
Moderate running exercise augments glycosaminoglycans and thickness of articu-
lar cartilage in the knee joint of young beagle dogs. J. Orthop. Res. 6, 188–195.
22. Helminen, H. J., Jurvelin, J., Kiviranta, I., Paukkonen, K., Säämänen, A.-M.,
and Tammi, M. (1987) Joint loading effects on articular cartilage: a historical
review, in Joint Loading: Biology and Health of Articular Structures (Helminen,
H. J., Kiviranta, I., Tammi, M., Säämänen, A.-M., Paukkonen, K,. and Jurvelin,
J., eds.), Wright, Bristol, UK, pp. 1–46.
23. Grodzinsky, A. J. (1983) Electromechanical and physicochemical properties of
connective tissue. CRC Crit. Rev. Bioeng. 9, 133–199.
24. Maroudas, A. (1979) Physico-chemical properties of articular cartilage, in Adult
Articular Cartilage 2nd ed., (Freeman, M. A. R., ed.), Pitman Medical,
Tunbridge Wells, England, UK, pp. 215–290.
25.
25 Mow, V. C., Wang, C. C., and Hung, C. T. (1999) The extracellular matrix,
interstitial fluid and ions as a mechanical signal transducer in articular cartilage.
26.
26 Urban, J. P. (2000) Present perspectives on cartilage and chondrocyte
mechanobiology. Biorheology 37, 185–190.
27.
27 Grodzinsky, A. J., Levenston, M. E., Jin, M., and Frank, E. H. (2000) Cartilage
tissue remodeling in response to mechanical forces. Annu. Rev. Biomed. Eng. 2,
691–713.
28. Guilak, F., Sah, R. L., and Setton, L. A. (1997) Physical regulation of cartilage
metabolism, in Basic Orthopaedic Biomechanics 2nd Ed. (Mow, V. C. and
Hayes, W. C., eds.), Raven, New York, NY, pp. 179–207.
29.
29 Burton-Wurster, N., Vernier-Singer, M., Farquhar, T., and Lust, G. (1993)
Effect of compressive loading and unloading on the synthesis of total protein,
proteoglycan, and fibronectin by canine cartilage explants. J. Orthop. Res. 11,
717–729.
30.
30 Gray, M. L., Pizzanelli, A. M., Grodzinsky, A. J., and Lee, R. C. (1988)
Mechanical and physicochemical determinants of the chondrocyte biosynthetic
response. J. Orthop. Res. 6, 777–792.
31.
31 Guilak, F., Meyer, B. C., Ratcliffe, A., and Mow, V. C. (1994) The effects of
matrix compression on proteoglycan metabolism in articular cartilage explants.
32.
32 Jones, I. L., Klamfeldt, D. D. S., and Sandstrom, T. (1982) The effect of continu-
ous mechanical pressure upon the turnover of articular cartilage proteoglycans
in vitro. Clin. Orthop. 165, 283–289.
33.
33 Sah, R. L., Kim, Y. J., Doong, J. H., Grodzinsky, A. J., Plaas, A. H. K., and
Sandy, J. D. (1989) Biosynthetic response of cartilage explants to dynamic com-
pression. J. Orthop. Res. 7, 619–636.
34 Copray, J. C. V. M., Jansen, H. W. B., and Duterloo, H. S. (1985) Effect of
34.
compressive forces on phosphatase activity in mandibular condylar cartilage of
the rat in vitro. J. Anat. 140, 479–489.
35.
35 Palmoski, M. J. and Brandt, K. D. (1984) Effects of static and cyclic com-
pressive loading on articular cartilage plugs in vitro. Arthritis Rheum. 27,
675–681.
36.
36 Parkkinen, J. J., Lammi, M. J., Helminen, H. J., and Tammi, M. (1992) Local
stimulation of proteoglycan synthesis in articular cartilage explants by dynamic
compression in vitro. J. Orthop. Res. 10, 610–620.
37.
37 Buschmann, M. D., Kim, Y. J., Wong, M., Frank, E., Hunziker, E. B., and
Grodzinsky, A. J. (1999) Stimulation of aggrecan synthesis in cartilage explants
by cyclic loading is localized to regions of high interstitial fluid flow. Arch.
Biochem. Biophys. 366, 1–7.
38.
38 Kim, Y. J., Sah, R. L., Grodzinsky, A. J., Plaas, A. H. K., and Sandy, J. D.
(1994) Mechanical regulation of cartilage biosynthetic behavior: physical
stimuli. Arch. Biochem. Biophys. 311, 1–12.
39.
39 Buschmann, M. D., Gluzband, Y. A., and Grodzinsky, A. J. (1995) Mechanical
compression modulates matrix biosynthesis in chondrocyte/agarose culture.
J. Cell Sci. 108, 1497–1508.
40.
40 Lee, D. A. and Bader, D. L. (1997) Compressive strains at physiological fre-
quencies influence the metabolism of chondrocytes seeded in agarose. J. Orthop.
Res. 15, 181–188.
41.
41 Ragan, P. M., Chin, V. I., Hung, H. H., et al. (2000) Chondrocyte extracellular
matrix synthesis and turnover are influenced by static compression in a new
alginate disk culture system. Arch. Biochem. Biophys. 383, 256–264.
346 Li et al.
42. Fukuda, K., Kumano, F., Asada, S., Saitoh, M., and Tanaka, S. (1997) Cyclic
tensile stretch loaded on bovine articular chondrocytes inhibits protein kinase C
activity. Osteoarthritis Cartilage 5, A38.
43.
43 Smith, R. L., Donlon, B. S., Gupta, M. K., et al. (1995) Effects of fluid-induced
shear on articular chondrocyte morphology and metabolism in vitro. J. Orthop.
Res. 13, 824–831.
44.
44 Parkkinen, J. J., Ikonen, J., Lammi, M. J., Laakkonen, J., Tammi, M., and
Helminen, H. J. (1993) Effects of cyclic hydrostatic pressure on proteoglycan
synthesis in cultured chondrocytes and articular cartilage explants. Arch.
Biochem. Biophys. 300, 458–465.
45.
45 Thibault, M., Poole, A. R., and Buschmann, M. D. (2002) Cyclic compression of
cartilage/bone explants in vitro leads to physical weakening, mechanical break-
down of collagen and release of matrix fragments. J. Orthop. Res. 20, 1265–1273.
46.
46 Loening, A., Levenston, M., James, I., Nuttal, M., et al. (2000) Injurious me-
chanical compression of bovine articular cartilage induces chondrocyte
apoptosis. Arch. Biochem. Biophys. 381, 205–212.
47. D’Lima, D. D., Hashimoto, S., Chen, P. C., Colwell, C. W. Jr., and Lotz, M. K.
(2001) Impact of mechanical trauma on matrix and cells. Clin. Orthop. 391
(Suppl.), S90–S99.
48 Kurz, B., Jin, M., Patwari, P., Cheng, D. M., Lark, M. W., and Grodzinsky, A. J.
48.
(2001) Biosynthetic response and mechanical properties of articular cartilage
after injurious compression. J. Orthop. Res. 19, 1140–1146.
49.
49 Clements, K. M., Bee, Z. C., Crossingham, G. V., Adams, M. A., and Sharif, M.
(2001) How severe must repetitive loading be to kill chondrocytes in articular
cartilage? Osteoarthritis Cartilage 9, 499–507.
50.
50 Quinn, T. M., Allen, R. G., Schalet, B. J., Perumbuli, P., and Hunziker, E. B.
(2001) Matrix and cell injury due to sub-impact loading of adult bovine articu-
lar cartilage explants: effects of strain rate and peak stress. J. Orthop. Res. 19,
242–249.
51. D’Lima, D. D., Hashimoto, S., Chen, P. C., Lotz, M. K., and Colwell, C. W. Jr.
(2001) Cartilage injury induces chondrocyte apoptosis. J. Bone Joint. Surg. Am.
83-A(Suppl. 2), 19–21.
52. Freed, L. E., Grande, D. A., Lingbin, Z., Emmanual, J., Marquis, J. C., and
Langer, R. (1994) Joint resurfacing using allograft chondrocytes and synthetic
biodegradable polymer scaffolds. J. Biomed. Mater. Res. 28, 891–899.
53. Amiel, D., Chu, C. R., Sah, R. L., and Coutts, R. D. (1998) Tissue engineering
of articular cartilage: perichondrial cells in osteochondral repair. Cells Mat. 8,
161–174.
54.
54 Nakahara, H., Goldberg, V. M., and Caplan, A. I. (1991) Culture-expanded
human periosteal-derived cells exhibit osteochondral potential in vivo. J. Orthop.
Res. 9, 465–476.
55. Wakitani, S., Goto, T., Pineda, S. J., et al. (1994) Mesenchymal cell-based repair
of large, full-thickness defects of articular cartilage. J. Bone Joint. Surg. 76-A,
579–592.
56.
56 Johnstone, B. and Yoo, J. U. (1999) Autologous mesenchymal progenitor cells
in articular cartilage repair. Clin. Orthop. 367S, 156–162.
57.
57 Zuk, P. A., Zhu, M., Mizuno, H., et al. (2001) Multilineage cells from human
adipose tissue: implications for cell-based therapies. Tissue Eng. 7, 211–228.
58. Häuselmann, H. J., Masuda, K., Hunziker, E. B., et al. (1996) Adult human
chondrocytes cultured in alginate form a matrix similar to native human articu-
lar cartilage. Am. J. Physiol. 40, C742–C752.
59. Chu, C. R., Coutts, R. D., Yoshioka, M., Harwood, F. L., Monosov, A. Z., and
Amiel, D. (1995) Articular cartilage repair using allogeneic perichondrocyte
seeded biodegradable porous polylactic acid (PLA): a tissue engineering study.
J. Biomed. Mater. Res. 29, 1147–1154.
60.
60 Moran, J. M., Pazzano, D., and Bonassar, L. J. (2003) Characterization of
polylactic acid-polyglycolic acid composites for cartilage tissue engineering.
Tissue Eng. 9, 63–70.
61. Ben-Yishay, A., Grande, D. A., Schwartz, R. E., Menche, D., and Pitman, M. D.
(1995) Repair of articular cartilage defects with collagen-chondrocyte allografts.
Tissue Eng. 1, 119–133.
62 Lee, C. R., Grodzinsky, A. J., Hsu, H. P., and Spector, M. (2003) Effects of a
62.
cultured autologous chondrocyte-seeded type II collagen scaffold on the healing
of a chondral defect in a canine model. J. Orthop. Res. 21, 272–281.
63.
63 Solchaga, L. A., Yoo, J. U., Lundberg, M., et al. (2000) Hyasluronan-
based polymers in the treatment of osteochondral defects. J. Orthop. Res. 18,
773–780.
64.
64 Lee, K. Y. and Mooney, D. J. (2001) Hydrogels for tissue engineering. Chem.
Rev. 101, 1869–1879.
65.
65 Mauck, R. L., Soltz, M. A., Wang, C. C., et al. (2000) Functional tissue engi-
neering of articular cartilage through dynamic loading of chondrocyte-seeded
agarose gels. J. Biomech. Eng. 122, 252–260.
66.
66 Guo, J. F., Jourdian, G. W., and MacCallum, D. K. (1989) Culture and growth
characteristics of chondrocytes encapsulated in alginate beads. Connect. Tissue
Res. 19, 277–297.
67.
67 van Susante, J. L., Buma, P., Schuman, L., Homminga, G. N., van den Berg,
W. B., and Veth, R. P. (1999) Resurfacing potential of heterologous
chondrocytes suspended in fibrin glue in large full-thickness defects of femo-
ral articular cartilage: an experimental study in the goat. Biomaterials 20,
1167–1175.
68.
68 Bryant, S. J. and Anseth, K. S. (2003) Controlling the spatial distribution of
ECM components in degradable PEG hydrogels for tissue engineering cartilage.
J. Biomed. Mater. Res. 64A, 70–79.
69.
69 Elisseeff, J., Anseth, K., Sims, D., McIntosh, W., Randolph, M., and Langer, R.
(1999) Transdermal photopolymerization for minimally invasive implantation.
Proc. Natl. Acad. Sci. USA 96, 3104–3107.
70.
70 Pittenger, M. F., Mackay, A. M., Beck, S. C., et al. (1999) Multilineage potential
of adult human mesenchymal stem cells. Science 284, 143–147.
348 Li et al.
71.
71 Kandel, R. A., Chen, H., Clark, J., and Renlund, R. (1995) Transplantation of
cartilagenous tissue generated in vitro into articular joint defects. Artif. Cells
Blood Substit. Immobil. Biotechnol. 23, 565–577.
72.
72 Williams, K. A., Saini, S., and Wick, T. M. (2002) Computational fluid dynam-
ics modeling of steady-state momentum and mass transport in a bioreactor for
cartilage tissue engineering. Biotechnol. Prog. 18, 951–963.
73.
73 Park, A., Wu, B., and Griffith, L. G. (1998) Integration of surface modifica-
tion and 3D fabrication techniques to prepare patterned poly(L-lactide) sub-
strates allowing regionally selective cell adhesion. J. Biomater. Sci. Polym.
Ed. 9, 89–110.
74 Chang, S. C., Rowley, J. A., Tobias, G., et al. (2001) Injection molding of chon-
74.
drocyte/alginate constructs in the shape of facial implants. J. Biomed. Mater.
Res. 55, 503–511.
75.
75 Khouri, R. K., Koudsi, B., and Reddi, H. (1991) Tissue transformation into bone
in vivo. A potential practical application. JAMA 266, 1953–1955.
76.
76 Pei, M., Solchaga, L. A., Seidel, J., et al. (2002) Bioreactors mediate the effec-
tiveness of tissue engineering scaffolds. FASEB J. 16, 1691–1694.
77.
77 Vunjak-Novakovic, G., Obradovic, B., Martin, I., Bursac, P. M., Langer, R., and
Freed, L. E. (1998) Dynamic cell seeding of polymer scaffolds for cartilage tis-
sue engineering. Biotechnol. Prog. 14, 193–202.
78 Martin, I., Obradovic, B., Treppo, S., et al. (2000) Modulation of the mechanical
78.
properties of tissue engineered cartilage. Biorheology 37, 141–147.
79. Ahsan, T., Chen, A. C., Chin, L., et al. (2003) Effects of long-term growth on
tissue engineered cartilage. Trans. Orthop. Res. Soc. 28, 309.
80. Freshney, R. I. (1994) Culture of Animal Cells: A Manual of Basic Technique
3rd. Ed., Wiley-Liss, New York, NY.
81.
81 Sah, R. L., Doong, J. Y. H., Grodzinsky, A. J., Plaas, A. H. K., and Sandy, J. D.
(1991) Effects of compression on the loss of newly synthesized proteoglycans
and proteins from cartilage explants. Arch. Biochem. Biophys. 286, 20–29.
82 Davisson, T. H., Sah, R. L., and Ratcliffe, A. R. (2002) Perfusion increases cell
82.
content and matrix synthesis in chondrocyte three-dimensional cultures. Tissue
Eng. 8, 807–816.
83.
83 Gray, M. L., Pizzanelli, A. M., Lee, R. C., Grodzinsky, A. J., and Swann, D. A.
(1989) Kinetics of the chondrocyte biosynthetic response to compressive load
and release. Biochim. Biophys. Acta 991, 415–425.
84. Kisiday, J., Siparsky, P. N., and Grodzinsky, A. J. (2003) Anabolic and cata-
bolic response to dynamic compression in a chondrocyte-seeded self-assembling
peptide hydrogel. Trans. Orthop. Res. Soc. 28, 304.
85. Kisiday, J., Jin, M., and Grodzinsky, A. J. (2002) Effects of dynamic compres-
sive loading duty cycle on in vitro conditioning of chondrocyte-seeded peptide
and agarose scaffolds. Trans. Orthop. Res. Soc. 27, 216.
86
86. Williamson, A. K., Chen, A. C., and Sah, R. L. (2001) Compressive properties
and function-composition relationships of developing bovine articular cartilage.
J. Orthop. Res. 19, 1113–1121.
87.
87 Quinn, T. M., Grodzinsky, A. J., Hunziker, E. B., and Sandy, J. D. (1998)
Effects of injurious compression on matrix turnover around individual cells in
calf articular cartilage explants. J. Orthop. Res. 16, 490–499.
88.
88 Jeffrey, J. E., Thomson, L. A., and Aspden, R. M. (1997) Matrix loss and syn-
thesis following a single impact load on articular cartilage in vitro. Biochim.
Biophys. Acta 1334, 223–232.
89.
89 Jin, M., Frank, E. H., Quinn, T. M., Hunziker, E. B., and Grodzinsky, A. J.
(2001) Tissue shear deformation stimulates proteoglycan and protein biosynthe-
sis in bovine cartilage explants. Arch. Biochem. Biophys. 395, 41–48.
90 Frank, E. H., Jin, M., Loening, A. M., Levenston, M. E., and Grodzinsky, A. J.
90.
(2000) A versatile shear and compression apparatus for mechanical stimulation
of tissue culture explants. J. Biomech. 33, 1523–1527.
91.
91 Vunjak-Novakovic, G., Martin, I., Obradovic, B., et al. (1999) Bioreactor culti-
vation conditions modulate the composition and mechanical properties of tis-
sue-engineered cartilage. J. Orthop. Res. 17, 130–139.
92.
92 Vunjak-Novakovic, G., Obradovic, B., Martin, I., and Freed, L. E. (2002)
Bioreactor studies of native and tissue engineered cartilage. Biorheology 39,
259–268.
93.
93 Obradovic, B., Meldon, J. H., Freed, L. E., and Vunjak-Novakovic, G. (2000)
Glycosaminoglycan deposition in engineered cartilage: experiments and math-
ematical model. AICHE J. 46, 1860–1871.
94. Davisson, T. H., Ratcliffe, A., and Sah, R. L. (2000) Flow-induced physical
stimuli during cartilage tissue engineering. Trans. Orthop. Res. Soc. 25, 610.
95. Davisson, T. H., Wu, F. J., Jain, D., Sah, R. L., and Ratcliffe, A. R. (1999) Effect
of perfusion on the growth of tissue engineered cartilage. Trans. Orthop. Res.
Soc. 24, 811.
96. Sittinger, M., Schultz, O., Keyszer, G., Minuth, W. W., and Burmester, G. R.
(1997) Artificial tissues in perfusion culture. Int. J. Artif. Organs 20, 57–62.
97.
97 Pazzano, D., Mercier, K. A., Moran, J. M., et al. (2000) Comparison of chon-
drogensis in static and perfused bioreactor culture. Biotechnol. Prog. 16, 893–896.
98. Davisson, T. H., Kunig, S., Chen, A. C., Sah, R. L., and Ratcliffe, A. (2002) The
effects of perfusion and compression on modulation of tissue engineered carti-
lage. Trans. Orthop. Res. Soc. 277, 488.
99.
99 Buschmann, M. D., Gluzband, Y. A., Grodzinsky, A. J., Kimura, J. H., and
Hunziker, E. B. (1992) Chondrocytes in agarose culture synthesize a mechani-
cally functional extracellular matrix. J. Orthop. Res. 10, 745–758.
100. Mok, S. S., Masuda, K., Häuselmann, H. J., Aydelotte, M. B., and Thonar, E. J.
(1994) Aggrecan synthesized by mature bovine chondrocytes suspended in algi-
nate. Identification of two distinct metabolic matrix pools. J. Biol. Chem. 269,
33,021–33,027.
101.
101 Giurea, A., Klein, T. J., Chen, A. C., et al. (2003) Adhesion of perichondrial
cells to a polylactic acid scaffold. J. Orthop. Res. 21, 584–589.
102
102. Bugbee, W. D. and Convery, F. R. (1999) Osteochondral allograft transplanta-
tion. Clin. Sports Med. 18, 67–75.
350 Li et al.
103. Outerbridge, H. K., Outerbridge, A. R., and Outerbridge, R. E. (1995) The use
of a lateral patellar autologous graft for the repair of a large osteochondral defect
in the knee. J. Bone Joint Surg. Am. 77-A, 65–72.
104.
104 McDermott, A. G., Langer, F., Pritzker, K. P., and Gross, A. E. (1985) Fresh
small-fragment osteochondral allografts. Long-term follow-up study on first 100
cases. Clin. Orthop. 197, 96–102.
105.
105 Yamashita, F., Sakakida, K., Suzu, F., and Takai, S. (1985) The transplantation
of an autogeneic osteochondral fragment for osteochondritis dissecans of the
knee. Clin. Orthop. 201, 43–50.
106 Bobic, V. (1996) Arthroscopic osteochondral autograft transplantation in ante-
106.
rior cruciate ligament reconstruction: a preliminary clinical study. Knee Surg.
Sports Traumatol. Arthrosc. 3, 262–264.
107.
107 Matsusue, Y., Yamamuro, T., and Hama, H. (1993) Arthroscopic multiple osteo-
chondral transplantation to the chondral defect in the knee associated with ante-
rior cruciate ligament disruption. Arthroscopy 9, 318–321.
108.
108 Benya, P. D. and Shaffer, J. D. (1982) Dedifferentiated chondrocytes reexpress
the differentiated collagen phenotype when cultured in agarose gels. Cell 30,
215–224.
109.
109 Bonaventure, J., Kadhom, N., Cohen-Solal, L., et al. (1994) Re-expression of
cartilage-specific genes by dedifferentiated human articular chondrocytes cul-
tured in alginate beads. Exp. Cell Res. 212, 97–104.
110 Chen, A. C. and Sah, R. L. (1998) The effect of static compression on
110.
proteoglycan synthesis by chondrocytes transplanted to articular cartilage in
vitro. J. Orthop. Res. 16, 542–550.
111. Li, K. W., Williamson, A. K., Wang, A. S., and Sah, R. L. (2001) Growth
responses of cartilage to static and dynamic compression. Clin. Orthop. 391S,
34–48.
112. Schreiber, R. E., Ilten-Kirby, B. M., Dunkelman, N. S., et al. (1999) Repair of
osteochondral defects with allogeneic tissue engineered cartilage implants. Clin.
Orthop. 367 (Suppl), 382–395.
113.
113 Klein, T. K., Schumacher, B. L., Schmidt, T. A., et al. (2003) Tissue engineering
of stratified articular cartilage from chondrocyte subpopulations. Osteoarthritis
Cartilage 11, 592–602.
114 Davisson, T. H., Kunig, S., Chen, A. C., Sah, R. L., and Ratcliffe, A. (2002)
114.
Static and dynamic compression modulate biosynthesis in tissue engineered car-
tilage. J. Orthop. Res. 20, 842–848.
115. Sah, R. L. (2003) The biomechanical faces of articular cartilage, in The Many
Faces of Osteoarthritis (Hascall, V. C., Kuettner, K. E., and Krall, A. M., eds.),
Birkhauser Verlag, Basel, Switzerland, pp. 506.
116. Bullough, P. G. and Cawston, T. E. (1994) Pathology and biochemistry of
osteoarthritis, in Color Atlas and Text of Osteoarthritis (Doherty, M., ed.), Times
Mirror International, London, UK, pp. 29–60.
117. Hunziker, E. B. (1992) Articular cartilage structure in humans and experimen-
tal animals, in Articular Cartilage and Osteoarthritis (Kuettner, K. E.,
Schleyerbach, R., Peyron, J. G., and Hascall, V. C., eds.), Raven, New York,
NY, pp. 183–199.
118. Rosenberg, L. C. and Buckwalter, J. A. (1986) Cartilage proteoglycans, in Arti-
cular Cartilage Biochemistry (Kuettner, K., Schleyerbach, R., and Hascall, V. C.,
eds.), Raven, New York, NY, pp. 39–57.
Index
ADAMTS, see Aggrecanase collagen immunoassays, 251–273

Adenoviral-mediated expression, 47 MAP kinases, 294–298, 303–304
Adenoviral transduction, see neoepitopes, 220–223, 237–249
Synoviocyte production, 244, 253, 257
Adenovirus, 40, 48, 147–162, 302 purification, 247, 253, 254, 260
Affinity purification, 237, 238, 247, use in flow cytometry, 183–208
260 Apoptosis, 275–289, 291–293, 303
Agarose annexin V, 277, 280–282
culture in, see Chondrocyte, caspase-3, 276, 279, 280, 285,
culture 293
gel, 70, 74–76, 80, 83, 103, 112, detection, 275–289
152,155 DNA strand breaks, 275, 282,
Aggrecan, 1, 16, 23, 30, 32, 44, 45, 283
62, 86, 111, 183, 191, 193– propidium iodide, 277, 280–282,
201, 237 284
antibody, 189, 210, 237–249 TUNEL, 277, 282–285, 287
degradation, 219–236, 237–249 Ascorbate, 42, 45,49, 55, 62, 190,
flow cytometry assay, 193–197 340
neoepitope, 220–222, 237–249
primers, see Gene expression Bifunctional reagent, 239
Aggrecanase, 219, 228, 233, 234, Bioinformatics, 122, 123, 175–181
241 Bioreactor, 310, 318, 328–332, 337–
inhibitor, 220–224 341
primers, see Gene, expression
Aggregate culture, 53 Cartilage
Alginate, 2, 15–22, 29, 37, 183, 184, apoptosis, 275–289
190, 191, 200, 307, 317 chondrocyte isolation from, 3–11,
bead, 15–22, 42, 190 17, 18, 157, 158, 167, 169,
culture, 15–22, 39, 42, 186, 203, 185, 187, 211, 212, 309, 314
330 degradation, 91, 219–236
Anabolism, 209–217 explant culture, 219–236, 251,
Annexin V, 275, 277, 280–283 336
Antibody gene expression analysis, 105–
caspase-3, 276, 279, 280 108, 109–125
characterization, 245, 254, 261 in situ hybridization, 105–108
353
354 Index
mechanical injury, 275, 283 morphology, 6, 29

mechanical stimulation, 325–351 necrosis, 276
mechanotransduction, 307, 308 phenotyping, 1–14
microarray analysis, 109–125 proliferation, 19, 20, 307, 313,
RNA extraction, 101–103, 111 319, 322
tissue engineering, 325–341 proteoglycan synthesis, 209–217,
Caspase-3, 276, 279, 280, 283, 285, 313
293 proteomics, 165–182
Cell line, see Chondrocyte sulphate incorporation, 209–217,
Cetyl pyridinium chloride, 209, 210, 313
213, 216 transfection, 129–145, 147–163
Chondrocyte Chondrogenesis, 53–67
anabolism, 209–217 Collagen, 1, 2, 16, 24, 32, 37, 42,
apoptosis, 275–289, 291–293 62, 69, 72, 77, 78, 85, 103,
cell line, 23–35, 37–52 108, 172, 184, 188, 189,
compressive strain, 307–324 191, 193–202, 219–236,
culture 251–273, 330
in 3D scaffolds, 39, 40, 42, degradation, 219–236, 257, 272
167, 172, 307–309, 314 flow cytometry, 193–198
in agarose, 37–41, 48, 183– hydroxyproline assay, 225, 226,
186, 190–198, 203, 307– 231–233
309, 314–323, 330, 333 neoepitope, 223, 251–273
in alginate, see Alginate primers, see Gene expression
`in monolayer, 1–14, 41, 166, type I, 2, 15, 16, 32, 37, 62, 167,
167, 169, 187, 211, 212 183, 188, 189, 193–198,
death, 275–289 253, 257, 258, 266, 270
differential gene expression, 109– type II, 1, 2, 16, 24, 32, 37, 49,
128 62, 69, 72, 77, 85, 103, 105,
differentiation, 2, 15, 16, 43, 53– 108, 183, 188, 189, 191,
67, 198, 203, 291, 292 193–202, 219, 221, 223,
electroporation, 129–145 231, 235, 251–273
flow cytometry, 183–208 type III, 2, 32, 37, 266, 270
gene expression, 69–78, 79–100, type V, 266
109–127 type VI, 266
gene therapy, 43 type IX, 1, 2, 221, 251–273
immortalization, 2, 23–35, 37–52 type X, 2, 38, 62, 108, 266
isolation, see Cartilage type XI, 1, 221
mechanical stimulation, 307–324, Collagenase
325–351 collagenase 1, see Matrix
microarray analysis, 109–128 metalloproteinase-1
Index 355
collagenase 2, see Matrix FuGENE 6 transfection, 131, 139–141

metalloproteinase-8
collagenase 3, see Matrix Gel electrophoresis, see SDS-PAGE
metalloproteinase-13 Gelatinase
for chondrocyte isolation, see -A, see Matrix metalloproteinase-2
Cartilage -B, see Matrix metalloproteinase-9
Coupling reagent, 238, 239, 253, Gene
255, 258, 263 expression
COX-2, see Cyclooxygenase-2 aggrecan primers, 86, 94
Culture, see Chondrocyte aggrecanase primers, 94
Cyclooxygenase-2, 43, 293, 299, collagen primers, 72, 86, 94,
301, 302 95
immunohistochemical detection, in situ hybridization, 105–108
299 matrix metalloproteinase
primers, 72, 95
Degradation microarray analysis, 109–128
of aggrecan, see Aggrecan RT-PCR, 28–32, 44, 45, 69–
of cartilage, see Cartilage 78, 79–100, 101–104, 159–
of collagen, see Collagen 161
of proteoglycans, see SOX9 primers, 81, 95
Proteoglycan TaqMan, 85–99
Differential gene expression, 109– therapy, 43
128 transfer, 24, 32, 147–163
Differentiation, see Chondrocyte GFAT, see Glutamine fructose-6-
DNA strand breaks, see Apoptosis phosphate amidotransferase
Glucosamine, 147, 148, 276–282,
Electroporation, 129–145 285
ERK, see MAP kinase Glutamine fructose-6-phosphate
Exponential decay electroporation, amido-transferase, 147–152,
131, 135–139 157–161
Extracellular matrix, 1, 20, 48, 62, Glycosaminoglycan, 209, 210,
69, 101, 111, 125, 183, 184, 247, 328
192–204, 219, 223, 252 Guanidine hydrochloride, 209, 211,
deformation, 308, 325, 331–334 213
material properties, 327, 334
Extraction of RNA from cartilage, HPLC, 237–240
see Cartilage Human papillomavirus type 16 early
function genes E6 and E7, 23,
Fibronectin, 183, 188, 191, 202 37, 40
Flow cytometry, 183–208, 281, 282 Hyaluronan, 32, 38, 183, 184, 210
356 Index
Hyaluronidase, 3–7, 23, 25, 28, 32, In vitro kinase assay, see MAP
32, 167, 169, 185,187, 204 kinase
Hydroxyproline assay, 223, 225, 226,
231–233 JNK, see MAP kinase
IGF-I, see Insulin-like growth factor Luciferase reporter assay, 40, 44,
IL-1, see Interleukin-1 46, 48, 50, 140, 141
IL-1RI, see Interleukin-1 receptor,
type I MALDI-TOF, 165, 166, 169, 174–
IL-1RII, see Interleukin-1 receptor, 180
type II MAP kinase, 291–305
Insulin-like growth factor activity, 295–299
IGF-I, 183, 188, 198, 201, 202, ERK, 292–294, 296–298, 300–
211, 214–216 304
IGF binding protein-3, 209, 211, immunohistochemistry, 297–299
214–216 immunofluorescence microscopy,
IGF receptor, type I, 183, 188, 297–299
198, 201, 202 in vitro kinase assay, 294, 296,
Interleukin-1, 2, 43, 46, 77, 121, 298
148, 159, 183, 188, 198– JNK, 292, 294, 296–298, 300–
203, 220–225, 229–231, 304
268, 291, 293, 295, 297, p38, 292–294, 296–298, 300–304
298, 301, 302 pharmacological inhibitors, 294,
receptor, type I, 183, 188, 198, 295, 300–303
201–203 Western blot analysis, 295–298,
receptor, type II, 183, 188, 198, 301
201–203 Matrix metalloproteinase
Immortalization, see Chondrocyte collagen degradation, see
Immunoassay Collagen
aggrecan, 197–199 inhibitor, 220–224, 269
type II collagen, 251–273 MMP-1, 44, 91, 92, 95, 115, 188,
type IX collagen, 251–273 221, 251, 257, 266, 268
Immunofluorescence microscopy, MMP-2, 115, 223, 252
see MAP kinase MMP-3, 44, 69, 72, 95, 188, 223
Immunohistochemistry, see MAP MMP-8, 91, 92, 95, 115, 221,
kinase 251, 268
In-gel protein digestion, 168, 171 MMP-9, 115, 223, 252
In situ hybridization, MMP-11, 115
type II collagen, 105–108 MMP-13, 43, 69, 72, 91, 92, 95,
type X collagen, 105–108 115, 189, 221, 251, 266, 268
Index 357
MMP-14, 95, 115, 251 PGE2, 291, 301

MMP-17, 115 Polybrene, 23, 26–28
primers, see Gene expression PolyHEMA, 37–41, 48
proteoglycan degradation, see Primer, see Gene, expression
Proteoglycan Proliferation, see Chondrocyte
Mechanical Propidium iodide, 187, 192, 195,
compression, 219, 277, 283, 287, 275, 277, 280–282, 284
307–324, 325–351 Protein kinase C, 291, 293
injury, see Cartilage, mechanical Proteoglycan
stimulation degradation, 202, 219–236, 237–
of cartilage, 325–351 249
of chondrocytes, 307–324, 325– precipitation, see Cetyl pyridinium
351 chloride
stress, 125, 200, 325, 334 synthesis, 209–217, 313
Mesenchymal stem cells, 53–67 Proteomics, 165–182
Microarray analysis, 109–128
MMP, see Matrix metalloproteinase Radio frequency electroporation,
130–135
Retroviral infection, 23, 26, 28
Necrosis, 276 Retrovirus production, 23, 26
Neoepitope RNA
aggrecan, 220–222, 237–249 amplification, see Gene,
type II collagen, 223, 251–273 expression
type IX collagen, 223, 251–273 extraction, see Cartilage
Nitric oxide, 148, 159, 291–293, in situ hybridization, 105–108
303, 304 quantification, 79–100
measurement, 159 RT-PCR, see Gene, expression
NO, see Nitric oxide RT-PCR, see Gene, expression
Nonviral transfection, see
Chondrocyte, transfection SDS-PAGE, 165–182, 237, 239,
Nuclear factor-κB, 293 243, 246, 259, 264, 270,
Nutridoma, 23, 25, 30, 37, 38,41, 279, 296–298
45–50 Shear stress, 325, 334
OH-Pro, see Hydroxyproline assay Simian virus 40 large T antigen, 23–
26, 28–30
p38, see MAP kinase SOX9, 43, 80–84, 95
Peptide-mass fingerprinting, 165, 166, Stromelysin-1, see Matrix
174–177 metalloproteinase, MMP-3
Pericellular matrix analysis, see Sulphate incorporation, 209–217,
Flow cytometry 313
358 Index
Synoviocyte TIMP, see Tissue inhibitor of

adenoviral transduction, 147, 148, metalloproteinase
150, 157–159 Tissue engineering, see Cartilage
culture, 158 Tissue inhibitor of
gene therapy, 147 metalloproteinase, 183, 189
gene transfer, 147 Transfection, 129–145, 147–163
isolation, 158 Transforming growth factor-β, 77,
183, 188
TaqMan, see Gene expression receptor, type II, 183, 188
Telomerase, 23–26, 29, 37 TUNEL, see Apoptosis
TGF-β, see Transforming growth Two-dimensional electrophoresis,
factor-β 165-182
Thymidine incorporation, 307, 313,
319, 322 Viral titer, 23, 27, 157

Cartilage and Osteoarthritis

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Cartilage and Osteoarthritis

Загружено:

Авторское право:

Доступные форматы

M E T H O D S I N M O L E C U L A R M E D I C I N E TM

John M. Walker, SERIES EDITOR

This publication is printed on acid-free paper. ∞

Cover design by Patricia F. Cleary.

Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

12 In Vitro Gene Transfer to Chondrocytes and Synovial Fibroblasts

1 Generation and Use of Transgenic Mice As Models of Osteoarthritis

13 High-Resolution Ultrasonography for Analysis

STEVEN C. GHIVIZZANI • Center for Molecular Orthopedics, Harvard Medical

MARTIN K. LOTZ • Division of Arthritis Research, The Scripps Research

Sylvie Thirion and Francis Berenbaum

2.2. Isolation and Culture of Rabbit

2.3. Isolation and Culture of Rat

2.4. Isolation and Culture of Newborn Mouse Rib Chondrocytes

2.5. Isolation and Culture of Mouse Primary Articular Chondrocytes

Fig. 2. Micrographs of different species of primary cultured chondrocytes. (A) Human

3.2. Isolation and Culture of Rabbit Primary Articular Chondrocytes

3.3. Isolation and Culture of Rat

3.3.2. Isolation and Culture of Chondrocytes

3.4. Isolation and Culture of Newborn Mouse Rib Chondrocytes

3.5. Isolation and Culture of Mouse Primary Articular Chondrocytes

Frédéric De Ceuninck, Christophe Lesur,

Key Words: Chondrocyte; cartilage; alginate; differentiation; cell culture.

To maintain chondrocytes in their original phenotype, various methods have

3.2. Encapsulation of Chondrocytes in Alginate Beads

Fig 2. (A) Morphological appearance of alginate beads with encapsulated rabbit

Fig 1. (opposite page) Schematic representation of the encapsulation of chondrocytes

Fig. 3. (A) Morphological appearance of an alginate bead with encapsulated human

3.3. Experimental Procedures: Recommendations

Key Words: Immortalized chondrocytes; type II collagen; aggrecan; retrovirus production;

may be useful as reproducible culture systems for evaluating chondrocyte func-

may be packaged in PA317 or 293GPG. SV40-TAg-containing vectors are

3.1. Retrovirus Production

3.1.2. Production of Retrovirus (24)

G418–resistant CFU/mL = no. of colonies

3.2. Retroviral Infection of Chondrocytes

3.3. Selection and Establishment of Immortalized Chondrocytes

Fig. 1. Morphologies of immortalized human chondrocytes in monolayer culture.

Telomerase activity in hTERT-immortalized cells is analyzed by the telomerase

3.4. Characterization of Established Chondrocyte Cell Lines

Fig. 2. Representative growth curves from established immortalized chondrocyte

4. Safety considerations in handling viral stocks: Special care should be taken to

Key Words: Immortalized chondrocytes; type II collagen; aggrecan; monolayer culture;

maintained as nonclonal populations and selected by passaging through

Louis, MO), or Intergen (Purchase, NY). Reserve serum testing, which is

b. BD™ Three Dimensional Collagen Composite Scaffold (BD Biosciences):

2.3. Immortalized Chondrocyte Cell Lines

3.1. Culture Systems for Immortalized Chondrocytes

ture conditions that maintain differentiated chondrocyte features or that permit

3.1.2. Fluid Suspension Cultures

4. To recover cells for direct experimental analysis, for redistribution in agarose- or

3.1.3. Alginate Bead Cultures (see Note 5)

in 1.5-mL tube. Freeze samples at –70°C and subject to two freeze/thaw

3.2. Immortalized Chondrocyte Cell Lines

Fig. 1. RT-PCR analysis of mRNAs expressed by immortalized human

Fig. 3. Regulation of COL2A1 promoter activity by interleukin-1β (IL-1β) in

Fig. 4. Transient expression of green fluorescent protein in immortalized human

3.2.2. Cotransfections Using Plasmid Vectors

3.2.3. Adenovirally Mediated Expression of Recombinant Proteins

4. Add 1 mL of serum-free medium containing adenovirus at 1:125 MOI. Incubate

3.2.4. Luciferase Assay

compared with housekeeping and cell growth-associated genes. For experiments,

activates transcription factors that are common to inflammatory responses and